Authors: Vatsal Baherwani, Yixuan Ren, Abhinav Shrivastava
Abstract: Text-to-video diffusion models synthesize temporal motion and spatial appearance through iterative denoising, yet how motion is encoded across timesteps remains poorly understood. Practitioners often exploit the empirical heuristic that early timesteps mainly shape motion and layout while later ones refine appearance, but this behavior has not been systematically characterized. In this work, we proxy motion encoding in video diffusion timesteps by the trade-off between appearance editing and motion preservation induced when injecting new conditions over specified timestep ranges, and characterize this proxy through a large-scale quantitative study. This protocol allows us to factor motion from appearance by quantitatively mapping how they compete along the denoising trajectory. Across diverse architectures, we consistently identify an early, motion-dominant regime and a later, appearance-dominant regime, yielding an operational motion-appearance boundary in timestep space. Building on this characterization, we simplify current one-shot motion customization paradigm by restricting training and inference to the motion-dominant regime, achieving strong motion transfer without auxiliary debiasing modules or specialized objectives. Our analysis turns a widely used heuristic into a spatiotemporal disentanglement principle, and our timestep-constrained recipe can serve as ready integration into existing motion transfer and editing methods.
Authors: Dawnena Key
Abstract: This paper presents a real-time American Sign Language (ASL) recognition system utilizing a hybrid deep learning architecture combining 3D Convolutional Neural Networks (3D CNN) with Long Short-Term Memory (LSTM) networks. The system processes webcam video streams to recognize word-level ASL signs, addressing communication barriers for over 70 million deaf and hard-of-hearing individuals worldwide. Our architecture leverages 3D convolutions to capture spatial-temporal features from video frames, followed by LSTM layers that model sequential dependencies inherent in sign language gestures. Trained on the WLASL dataset (2,000 common words), ASL-LEX lexical database (~2,700 signs), and a curated set of 100 expert-annotated ASL signs, the system achieves F1-scores ranging from 0.71 to 0.99 across sign classes. The model is deployed on AWS infrastructure with edge deployment capability on OAK-D cameras for real-time inference. We discuss the architecture design, training methodology, evaluation metrics, and deployment considerations for practical accessibility applications.
Authors: Hassan Khalid, Muhammad Mahad Khaliq, Muhammad Jawad Bashir
Abstract: The rapid evolution of Artificial intelligence in healthcare has opened avenues for enhancing various processes, including medical billing and transcription. This paper introduces an innovative approach by integrating AI with Locally Linear Embedding (LLE) to revolutionize the handling of high-dimensional medical data. This AI-enhanced LLE model is specifically tailored to improve the accuracy and efficiency of medical billing systems and transcription services. By automating these processes, the model aims to reduce human error and streamline operations, thereby facilitating faster and more accurate patient care documentation and financial transactions. This paper provides a comprehensive mathematical model of AI-enhanced LLE, demonstrating its application in real-world healthcare scenarios through a series of experiments. The results indicate a significant improvement in data processing accuracy and operational efficiency. This study not only underscores the potential of AI-enhanced LLE in medical data analysis but also sets a foundation for future research into broader healthcare applications.
Authors: Zhaonan Li, Shijie Lu, Fei Wang, Jacob Dineen, Xiao Ye, Zhikun Xu, Siyi Liu, Young Min Cho, Bangzheng Li, Daniel Chang, Kenny Nguyen, Qizheng Yang, Muhao Chen, Ben Zhou
Abstract: End-to-end Vision-language Models (VLMs) often answer visual questions by exploiting spurious correlations instead of causal visual evidence, and can become more shortcut-prone when fine-tuned. We introduce VISTA (Visual-Information Separation for Text-based Analysis), a modular framework that decouples perception from reasoning via an explicit information bottleneck. A frozen VLM sensor is restricted to short, objective perception queries, while a text-only LLM reasoner decomposes each question, plans queries, and aggregates visual facts in natural language. This controlled interface defines a reward-aligned environment for training unbiased visual reasoning with reinforcement learning. Instantiated with Qwen2.5-VL and Llama3.2-Vision sensors, and trained with GRPO from only 641 curated multi-step questions, VISTA significantly improves robustness to real-world spurious correlations on SpuriVerse (+16.29% with Qwen-2.5-VL-7B and +6.77% with Llama-3.2-Vision-11B), while remaining competitive on MMVP and a balanced SeedBench subset. VISTA transfers robustly across unseen VLM sensors and is able to recognize and recover from VLM perception failures. Human analysis further shows that VISTA's reasoning traces are more neutral, less reliant on spurious attributes, and more explicitly grounded in visual evidence than end-to-end VLM baselines.
Authors: Antara Titikhsha, Divyanshu Tak
Abstract: Effective aneurysm detection is essential to avert life-threatening hemorrhages, but it remains challenging due to the subtle morphology of the aneurysm, pronounced class imbalance, and the scarcity of annotated data. We introduce SAMM2D, a dual-encoder framework that achieves an AUC of 0.686 on the RSNA intracranial aneurysm dataset; an improvement of 32% over the clinical baseline. In a comprehensive ablation across six augmentation regimes, we made a striking discovery: any form of data augmentation degraded performance when coupled with a strong pretrained backbone. Our unaugmented baseline model outperformed all augmented variants by 1.75--2.23 percentage points (p < 0.01), overturning the assumption that "more augmentation is always better" in low-data medical settings. We hypothesize that ImageNet-pretrained features already capture robust invariances, rendering additional augmentations both redundant and disruptive to the learned feature manifold. By calibrating the decision threshold, SAMM2D reaches 95% sensitivity, surpassing average radiologist performance, and translates to a projected \$13.9M in savings per 1,000 patients in screening applications. Grad-CAM visualizations confirm that 85% of true positives attend to relevant vascular regions (62% IoU with expert annotations), demonstrating the model's clinically meaningful focus. Our results suggest that future medical imaging workflows could benefit more from strong pretraining than from increasingly complex augmentation pipelines.
Authors: Xitong Ling, Minxi Ouyang, Xiaoxiao Li, Jiawen Li, Ying Chen, Yuxuan Sun, Xinrui Chen, Tian Guan, Xiaoping Liu, Yonghong He
Abstract: Multiple Instance Learning (MIL) has enabled weakly supervised analysis of whole-slide images (WSIs) in computational pathology. However, traditional MIL approaches often lose crucial contextual information, while transformer-based variants, though more expressive, suffer from quadratic complexity and redundant computations. To address these limitations, we propose HookMIL, a context-aware and computationally efficient MIL framework that leverages compact, learnable hook tokens for structured contextual aggregation. These tokens can be initialized from (i) key-patch visual features, (ii) text embeddings from vision-language pathology models, and (iii) spatially grounded features from spatial transcriptomics-vision models. This multimodal initialization enables Hook Tokens to incorporate rich textual and spatial priors, accelerating convergence and enhancing representation quality. During training, Hook tokens interact with instances through bidirectional attention with linear complexity. To further promote specialization, we introduce a Hook Diversity Loss that encourages each token to focus on distinct histopathological patterns. Additionally, a hook-to-hook communication mechanism refines contextual interactions while minimizing redundancy. Extensive experiments on four public pathology datasets demonstrate that HookMIL achieves state-of-the-art performance, with improved computational efficiency and interpretability. Codes are available at https://github.com/lingxitong/HookMIL.
Authors: Kenneth Xu, Songhan Wu
Abstract: The Segment Anything Model (SAM) enables promptable, high-quality segmentation but is often too computationally expensive for latency-critical settings. TinySAM is a lightweight, distilled SAM variant that preserves strong zero-shot mask quality, yet its "segment-everything" mode still requires hundreds of prompts and remains slow in practice. We first replicate TinySAM on COCO val2017 using official checkpoints, matching the reported AP within 0.03%, establishing a reliable experimental baseline. Building on this, we propose Tiny-YOLOSAM, a fast hybrid pipeline that uses a recent YOLO detector (YOLOv12) to generate box prompts for TinySAM on salient foreground objects, and supplements uncovered regions with sparse point prompts sampled only where YOLO-guided masks provide no coverage. On COCO val2017, the hybrid system substantially improves class-agnostic coverage (AR from 16.4% to 77.1%, mIoU from 19.2% to 67.8%) while reducing end-to-end runtime from 49.20s/image to 10.39s/image (4.7x) on an Apple M1 Pro CPU. These results suggest detector-guided prompting combined with targeted sparse sampling as an effective alternative to dense "segment-everything" prompting for practical full-scene segmentation.
Authors: Shivum Telang
Abstract: Diabetic Retinopathy (DR) is a leading cause of vision loss worldwide, requiring early detection to preserve sight. Limited access to physicians often leaves DR undiagnosed. To address this, AI models utilize lesion segmentation for interpretability; however, manually annotating lesions is impractical for clinicians. Physicians require a model that explains the reasoning for classifications rather than just highlighting lesion locations. Furthermore, current models are one-dimensional, relying on a single imaging modality for explainability and achieving limited effectiveness. In contrast, a quantitative-detection system that identifies individual DR lesions in natural language would overcome these limitations, enabling diverse applications in screening, treatment, and research settings. To address this issue, this paper presents a novel multimodal explainability model utilizing a VLM with few-shot learning, which mimics an ophthalmologist's reasoning by analyzing lesion distributions within retinal quadrants for fundus images. The model generates paired Grad-CAM heatmaps, showcasing individual neuron weights across both OCT and fundus images, which visually highlight the regions contributing to DR severity classification. Using a dataset of 3,000 fundus images and 1,000 OCT images, this innovative methodology addresses key limitations in current DR diagnostics, offering a practical and comprehensive tool for improving patient outcomes.
Authors: Qiang Guo, Rubo Zhang, Bingbing Zhang, Junjie Liu, Jianqing Liu
Abstract: Crowd counting typically relies on labor-intensive point-level annotations and computationally intensive backbones, restricting its scalability and deployment in resource-constrained environments. To address these challenges, this paper proposes the TCFormer, a tiny, ultra-lightweight, weakly-supervised transformer-based crowd counting framework with only 5 million parameters that achieves competitive performance. Firstly, a powerful yet efficient vision transformer is adopted as the feature extractor, the global context-aware capabilities of which provides semantic meaningful crowd features with a minimal memory footprint. Secondly, to compensate for the lack of spatial supervision, we design a feature aggregation mechanism termed the Learnable Density-Weighted Averaging module. This module dynamically re-weights local tokens according to predicted density scores, enabling the network to adaptively modulate regional features based on their specific density characteristics without the need for additional annotations. Furthermore, this paper introduces a density-level classification loss, which discretizes crowd density into distinct grades, thereby regularizing the training process and enhancing the model's classification power across varying levels of crowd density. Therefore, although TCformer is trained under a weakly-supervised paradigm utilizing only image-level global counts, the joint optimization of count and density-level losses enables the framework to achieve high estimation accuracy. Extensive experiments on four benchmarks including ShanghaiTech A/B, UCF-QNRF, and NWPU datasets demonstrate that our approach strikes a superior trade-off between parameter efficiency and counting accuracy and can be a good solution for crowd counting tasks in edge devices.
Authors: Md. Ismiel Hossen Abir, Awolad Hossain
Abstract: Malaria remains a prevalent health concern in regions with tropical and subtropical climates. The cause of malaria is the Plasmodium parasite, which is transmitted through the bites of infected female Anopheles mosquitoes. Traditional diagnostic methods, such as microscopic blood smear analysis, are low in sensitivity, depend on expert judgment, and require resources that may not be available in remote settings. To overcome these limitations, this study proposes a deep learning-based approach utilizing a custom Convolutional Neural Network (CNN) to automatically classify blood cell images as parasitized or uninfected. The model achieves an accuracy of 96%, with precision and recall scores exceeding 0.95 for both classes. This study also compares the custom CNN with established deep learning architectures, including ResNet50, VGG16, MobileNetV2, and DenseNet121. To enhance model interpretability, Explainable AI techniques such as SHAP, LIME, and Saliency Maps are applied. The proposed system shows how deep learning can provide quick, accurate and understandable malaria diagnosis, especially in areas with limited resources.
Authors: Naichuan Zheng, Xiahai Lun, Weiyi Li, Yuchen Du
Abstract: Graph Convolutional Networks (GCNs) demonstrate strong capability in modeling skeletal topology for action recognition, yet their dense floating-point computations incur high energy costs. Spiking Neural Networks (SNNs), characterized by event-driven and sparse activation, offer energy efficiency but remain limited in capturing coupled temporal-frequency and topological dependencies of human motion. To bridge this gap, this article proposes Signal-SGN++, a topology-aware spiking graph framework that integrates structural adaptivity with time-frequency spiking dynamics. The network employs a backbone composed of 1D Spiking Graph Convolution (1D-SGC) and Frequency Spiking Convolution (FSC) for joint spatiotemporal and spectral feature extraction. Within this backbone, a Topology-Shift Self-Attention (TSSA) mechanism is embedded to adaptively route attention across learned skeletal topologies, enhancing graph-level sensitivity without increasing computational complexity. Moreover, an auxiliary Multi-Scale Wavelet Transform Fusion (MWTF) branch decomposes spiking features into multi-resolution temporal-frequency representations, wherein a Topology-Aware Time-Frequency Fusion (TATF) unit incorporates structural priors to preserve topology-consistent spectral fusion. Comprehensive experiments on large-scale benchmarks validate that Signal-SGN++ achieves superior accuracy-efficiency trade-offs, outperforming existing SNN-based methods and achieving competitive results against state-of-the-art GCNs under substantially reduced energy consumption.
Authors: Abdellah Zakaria Sellam, Salah Eddine Bekhouche, Fadi Dornaika, Cosimo Distante, Abdenour Hadid
Abstract: Pedestrian Attribute Recognition (PAR) involves predicting fine-grained attributes such as clothing color, gender, and accessories from pedestrian imagery, yet is hindered by severe class imbalance, intricate attribute co-dependencies, and domain shifts. We introduce VLM-PAR, a modular vision-language framework built on frozen SigLIP 2 multilingual encoders. By first aligning image and prompt embeddings via refining visual features through a compact cross-attention fusion, VLM-PAR achieves significant accuracy improvement on the highly imbalanced PA100K benchmark, setting a new state-of-the-art performance, while also delivering significant gains in mean accuracy across PETA and Market-1501 benchmarks. These results underscore the efficacy of integrating large-scale vision-language pretraining with targeted cross-modal refinement to overcome imbalance and generalization challenges in PAR.
Authors: Hieu Minh Nguyen, Tam Le-Thanh Dang, Kiet Van Nguyen
Abstract: Understanding signboard text in natural scenes is essential for real-world applications of Visual Question Answering (VQA), yet remains underexplored, particularly in low-resource languages. We introduce ViSignVQA, the first large-scale Vietnamese dataset designed for signboard-oriented VQA, which comprises 10,762 images and 25,573 question-answer pairs. The dataset captures the diverse linguistic, cultural, and visual characteristics of Vietnamese signboards, including bilingual text, informal phrasing, and visual elements such as color and layout. To benchmark this task, we adapted state-of-the-art VQA models (e.g., BLIP-2, LaTr, PreSTU, and SaL) by integrating a Vietnamese OCR model (SwinTextSpotter) and a Vietnamese pretrained language model (ViT5). The experimental results highlight the significant role of the OCR-enhanced context, with F1-score improvements of up to 209% when the OCR text is appended to questions. Additionally, we propose a multi-agent VQA framework combining perception and reasoning agents with GPT-4, achieving 75.98% accuracy via majority voting. Our study presents the first large-scale multimodal dataset for Vietnamese signboard understanding. This underscores the importance of domain-specific resources in enhancing text-based VQA for low-resource languages. ViSignVQA serves as a benchmark capturing real-world scene text characteristics and supporting the development and evaluation of OCR-integrated VQA models in Vietnamese.
Authors: Tasha Pais, Nikhilesh Belulkar
Abstract: Semantic Abstraction's key observation is that 2D VLMs' relevancy activations roughly correspond to their confidence of whether and where an object is in the scene. Thus, relevancy maps are treated as "abstract object" representations. We use this framework for learning 3D localization and completion for the exclusive domain of hidden objects, defined as objects that cannot be directly identified by a VLM because they are at least partially occluded. This process of localizing hidden objects is a form of unstructured search that can be performed more efficiently using historical data of where an object is frequently placed. Our model can accurately identify the complete 3D location of a hidden object on the first try significantly faster than a naive random search. These extensions to semantic abstraction hope to provide household robots with the skills necessary to save time and effort when looking for lost objects.
Authors: Naishan Zheng, Jie Huang, Qingpei Guo, Feng Zhao
Abstract: Understanding long videos with multimodal large language models (MLLMs) remains challenging due to the heavy redundancy across frames and the need for temporally coherent representations. Existing static strategies, such as sparse sampling, frame compression, and clustering, are optimized for offline settings and often produce fragmented or over-compressed outputs when applied to continuous video streams. We present VideoScaffold, a dynamic representation framework designed for streaming video understanding. It adaptively adjusts event granularity according to video duration while preserving fine-grained visual semantics. VideoScaffold introduces two key components: Elastic-Scale Event Segmentation (EES), which performs prediction-guided segmentation to dynamically refine event boundaries, and Hierarchical Event Consolidation (HEC), which progressively aggregates semantically related segments into multi-level abstractions. Working in concert, EES and HEC enable VideoScaffold to transition smoothly from fine-grained frame understanding to abstract event reasoning as the video stream unfolds. Extensive experiments across both offline and streaming video understanding benchmarks demonstrate that VideoScaffold achieves state-of-the-art performance. The framework is modular and plug-and-play, seamlessly extending existing image-based MLLMs to continuous video comprehension. The code is available at https://github.com/zheng980629/VideoScaffold.
Authors: HaoNan Tang
Abstract: Vision Transformers (ViT) have demonstrated significant promise in dense prediction tasks such as pose estimation. However, their performance is frequently constrained by the overly simplistic front-end designs employed in models like ViTPose. This naive patchification mechanism struggles to effectively handle multi-scale variations and results in irreversible information loss during the initial feature extraction phase. To overcome this limitation, we introduce a novel KAN-enhanced FPN-Stem architecture. Through rigorous ablation studies, we first identified that the true bottleneck for performance improvement lies not in plug-and-play attention modules (e.g., CBAM), but in the post-fusion non-linear smoothing step within the FPN. Guided by this insight, our core innovation is to retain the classic "upsample-and-add" fusion stream of the FPN, but replace its terminal, standard linear 3x3 smoothing convolution with a powerful KAN-based convolutional layer. Leveraging its superior non-linear modeling capabilities, this KAN-based layer adaptively learns and rectifies the "artifacts" generated during the multi-scale fusion process. Extensive experiments on the COCO dataset demonstrate that our KAN-FPN-Stem achieves a significant performance boost of up to +2.0 AP over the lightweight ViTPose-S baseline. This work not only delivers a plug-and-play, high-performance module but, more importantly, reveals that: the performance bottleneck in ViT front-end often lies not in 'feature refinement' (Attention), but in the quality of 'feature fusion' (Fusion). Furthermore, it provides an effective path to address this bottleneck through the introduction of the KAN operator.
Authors: Mengxiao Geng, Ran Hong, Xiaoling Xu, Bingxuan Li, Qiegen Liu
Abstract: Low-dose PET imaging is crucial for reducing patient radiation exposure but faces challenges like noise interference, reduced contrast, and difficulty in preserving physiological details. Existing methods often neglect both projection-domain physics knowledge and patient-specific meta-information, which are critical for functional-semantic correlation mining. In this study, we introduce a meta-information guided cross-domain synergistic diffusion model (MiG-DM) that integrates comprehensive cross-modal priors to generate high-quality PET images. Specifically, a meta-information encoding module transforms clinical parameters into semantic prompts by considering patient characteristics, dose-related information, and semi-quantitative parameters, enabling cross-modal alignment between textual meta-information and image reconstruction. Additionally, the cross-domain architecture combines projection-domain and image-domain processing. In the projection domain, a specialized sinogram adapter captures global physical structures through convolution operations equivalent to global image-domain filtering. Experiments on the UDPET public dataset and clinical datasets with varying dose levels demonstrate that MiG-DM outperforms state-of-the-art methods in enhancing PET image quality and preserving physiological details.
Authors: Phi-Hung Hoang, Nam-Thuan Trinh, Van-Manh Tran, Thi-Thu-Hong Phan
Abstract: Deploying deep learning models on resource-constrained edge devices remains a major challenge in smart agriculture due to the trade-off between computational efficiency and recognition accuracy. To address this challenge, this study proposes a hybrid knowledge distillation framework for developing a lightweight yet high-performance convolutional neural network. The proposed approach designs a customized student model that combines inverted residual blocks with dense connectivity and trains it under the guidance of a ResNet18 teacher network using a multi-objective strategy that integrates hard-label supervision, feature-level distillation, response-level distillation, and self-distillation. Experiments are conducted on a rice seed variety identification dataset containing nine varieties and further extended to four plant leaf disease datasets, including rice, potato, coffee, and corn, to evaluate generalization capability. On the rice seed variety classification task, the distilled student model achieves an accuracy of 98.56%, which is only 0.09% lower than the teacher model (98.65%), while requiring only 0.68 GFLOPs and approximately 1.07 million parameters. This corresponds to a reduction of about 2.7 times in computational cost and more than 10 times in model size compared with the ResNet18 teacher model. In addition, compared with representative pretrained models, the proposed student reduces the number of parameters by more than 6 times relative to DenseNet121 and by over 80 times compared with the Vision Transformer (ViT) architecture, while maintaining comparable or superior classification accuracy. Consistent performance gains across multiple plant leaf disease datasets further demonstrate the robustness, efficiency, and strong deployment potential of the proposed framework for hardware-limited smart agriculture systems.
Authors: Aahan Sachdeva, Dhanvinkumar Ganeshkumar, James E. Gallagher, Tyler Treat, Edward J. Oughton
Abstract: Autonomous robotic platforms are playing a growing role across the emergency services sector, supporting missions such as search and rescue operations in disaster zones and reconnaissance. However, traditional red-green-blue (RGB) detection pipelines struggle in low-light environments, and thermal-based systems lack color and texture information. To overcome these limitations, we present an adaptive framework that fuses RGB and long-wave infrared (LWIR) video streams at multiple fusion ratios and dynamically selects the optimal detection model for each illumination condition. We trained 33 You Only Look Once (YOLO) models on over 22,000 annotated images spanning three light levels: no-light (<10 lux), dim-light (10-1000 lux), and full-light (>1000 lux). To integrate both modalities, fusion was performed by blending aligned RGB and LWIR frames at eleven ratios, from full RGB (100/0) to full LWIR (0/100) in 10% increments. Evaluation showed that the best full-light model (80/20 RGB-LWIR) and dim-light model (90/10 fusion) achieved 92.8% and 92.0% mean confidence; both significantly outperformed the YOLOv5 nano (YOLOv5n) and YOLOv11 nano (YOLOv11n) baselines. Under no-light conditions, the top 40/60 fusion reached 71.0%, exceeding baselines though not statistically significant. Adaptive RGB-LWIR fusion improved detection confidence and reliability across all illumination conditions, enhancing autonomous robotic vision performance.
Authors: Antara Titikhsha, Om Kulkarni, Dharun Muthaiah
Abstract: Text-to-image diffusion models generate highly detailed textures, yet they often rely on surface appearance and fail to follow strict geometric constraints, particularly when those constraints conflict with the style implied by the text prompt. This reflects a broader semantic gap between human perception and current generative models. We investigate whether geometric understanding can be introduced without specialized training by using lightweight, off-the-shelf discriminators as external guidance signals. We propose a Human Perception Embedding (HPE) teacher trained on the THINGS triplet dataset, which captures human sensitivity to object shape. By injecting gradients from this teacher into the latent diffusion process, we show that geometry and style can be separated in a controllable manner. We evaluate this approach across three architectures: Stable Diffusion v1.5 with a U-Net backbone, the flow-matching model SiT-XL/2, and the diffusion transformer PixArt-{\Sigma}. Our experiments reveal that flow models tend to drift back toward their default trajectories without continuous guidance, and we demonstrate zero-shot transfer of complex three-dimensional shapes, such as an Eames chair, onto conflicting materials such as pink metal. This guided generation improves semantic alignment by about 80 percent compared to unguided baselines. Overall, our results show that small teacher models can reliably guide large generative systems, enabling stronger geometric control and broadening the creative range of text-to-image synthesis.
Authors: Leslie Gu, Junhwa Hur, Charles Herrmann, Fangneng Zhan, Todd Zickler, Deqing Sun, Hanspeter Pfister
Abstract: We introduce GeCo, a geometry-grounded metric for jointly detecting geometric deformation and occlusion-inconsistency artifacts in static scenes. By fusing residual motion and depth priors, GeCo produces interpretable, dense consistency maps that reveal these artifacts. We use GeCo to systematically benchmark recent video generation models, uncovering common failure modes, and further employ it as a training-free guidance loss to reduce deformation artifacts during video generation.
Authors: Dingyu Wang, Zimu Yuan, Jiajun Liu, Shanggui Liu, Nan Zhou, Tianxing Xu, Di Huang, Dong Jiang
Abstract: Background: The rapid integration of foundation models into clinical practice and public health necessitates a rigorous evaluation of their true clinical reasoning capabilities beyond narrow examination success. Current benchmarks, typically based on medical licensing exams or curated vignettes, fail to capture the integrated, multimodal reasoning essential for real-world patient care. Methods: We developed the Bones and Joints (B&J) Benchmark, a comprehensive evaluation framework comprising 1,245 questions derived from real-world patient cases in orthopedics and sports medicine. This benchmark assesses models across 7 tasks that mirror the clinical reasoning pathway, including knowledge recall, text and image interpretation, diagnosis generation, treatment planning, and rationale provision. We evaluated eleven vision-language models (VLMs) and six large language models (LLMs), comparing their performance against expert-derived ground truth. Results: Our results demonstrate a pronounced performance gap between task types. While state-of-the-art models achieved high accuracy, exceeding 90%, on structured multiple-choice questions, their performance markedly declined on open-ended tasks requiring multimodal integration, with accuracy scarcely reaching 60%. VLMs demonstrated substantial limitations in interpreting medical images and frequently exhibited severe text-driven hallucinations, often ignoring contradictory visual evidence. Notably, models specifically fine-tuned for medical applications showed no consistent advantage over general-purpose counterparts. Conclusions: Current artificial intelligence models are not yet clinically competent for complex, multimodal reasoning. Their safe deployment should currently be limited to supportive, text-based roles. Future advancement in core clinical tasks awaits fundamental breakthroughs in multimodal integration and visual understanding.
Authors: Hussain Alasmawi, Numan Saeed, Mohammad Yaqub
Abstract: The growing demand for prenatal ultrasound imaging has intensified a global shortage of trained sonographers, creating barriers to essential fetal health monitoring. Deep learning has the potential to enhance sonographers' efficiency and support the training of new practitioners. Vision-Language Models (VLMs) are particularly promising for ultrasound interpretation, as they can jointly process images and text to perform multiple clinical tasks within a single framework. However, despite the expansion of VLMs, no standardized benchmark exists to evaluate their performance in fetal ultrasound imaging. This gap is primarily due to the modality's challenging nature, operator dependency, and the limited public availability of datasets. To address this gap, we present Fetal-Gauge, the first and largest visual question answering benchmark specifically designed to evaluate VLMs across various fetal ultrasound tasks. Our benchmark comprises over 42,000 images and 93,000 question-answer pairs, spanning anatomical plane identification, visual grounding of anatomical structures, fetal orientation assessment, clinical view conformity, and clinical diagnosis. We systematically evaluate several state-of-the-art VLMs, including general-purpose and medical-specific models, and reveal a substantial performance gap: the best-performing model achieves only 55\% accuracy, far below clinical requirements. Our analysis identifies critical limitations of current VLMs in fetal ultrasound interpretation, highlighting the urgent need for domain-adapted architectures and specialized training approaches. Fetal-Gauge establishes a rigorous foundation for advancing multimodal deep learning in prenatal care and provides a pathway toward addressing global healthcare accessibility challenges. Our benchmark will be publicly available once the paper gets accepted.
Authors: Philip Xu, David Elizondo, Raouf Hamzaoui
Abstract: We introduce Uni4D, a unified framework for large scale open vocabulary 3D retrieval and controlled 4D generation based on structured three level alignment across text, 3D models, and image modalities. Built upon the Align3D 130 dataset, Uni4D employs a 3D text multi head attention and search model to optimize text to 3D retrieval through improved semantic alignment. The framework further strengthens cross modal alignment through three components: precise text to 3D retrieval, multi view 3D to image alignment, and image to text alignment for generating temporally consistent 4D assets. Experimental results demonstrate that Uni4D achieves high quality 3D retrieval and controllable 4D generation, advancing dynamic multimodal understanding and practical applications.
Authors: Tian Guo, Hui Yuan, Philip Xu, David Elizondo
Abstract: We propose SirenPose, a novel loss function that combines the periodic activation properties of sinusoidal representation networks with geometric priors derived from keypoint structures to improve the accuracy of dynamic 3D scene reconstruction. Existing approaches often struggle to maintain motion modeling accuracy and spatiotemporal consistency in fast moving and multi target scenes. By introducing physics inspired constraint mechanisms, SirenPose enforces coherent keypoint predictions across both spatial and temporal dimensions. We further expand the training dataset to 600,000 annotated instances to support robust learning. Experimental results demonstrate that models trained with SirenPose achieve significant improvements in spatiotemporal consistency metrics compared to prior methods, showing superior performance in handling rapid motion and complex scene changes.
Authors: Vesal Ahsani, Babak Hossein Khalaj
Abstract: In-cabin Driver Monitoring Systems (DMS) must recognize distraction- and drowsiness-related behaviors with low latency under strict constraints on compute, power, and cost. We present a single-camera in-cabin driver behavior recognition system designed for deployment on two low-cost edge platforms: Raspberry Pi 5 (CPU-only) and Google Coral Edge TPU. The proposed pipeline combines (i) a compact per-frame vision model, (ii) a confounder-aware label design to reduce visually similar false positives, and (iii) a temporal decision head that triggers alerts only when predictions are both confident and sustained. The system covers 17 behavior classes, including multiple phone-use modes, eating/drinking, smoking, reaching behind, gaze/attention shifts, passenger interaction, grooming, control-panel interaction, yawning, and eyes-closed sleep. Training and evaluation use licensed datasets spanning diverse drivers, vehicles, and lighting conditions (details in Section 6), and we further validate runtime behavior in real in-vehicle tests. The optimized deployments achieve about 16 FPS on Raspberry Pi 5 with INT8 inference (per-frame latency under 60 ms) and about 25 FPS on Coral Edge TPU, enabling real-time monitoring and stable alert generation on inexpensive hardware. Finally, we discuss how reliable in-cabin human-state perception can serve as an upstream input for human-centered vehicle intelligence, including emerging agentic vehicle concepts.
Authors: Noor Fatima, Hasan Faraz Khan, Muzammil Behzad
Abstract: This work presents an attack-aware deepfake and image-forensics detector designed for robustness, well-calibrated probabilities, and transparent evidence under realistic deployment conditions. The method combines red-team training with randomized test-time defense in a two-stream architecture, where one stream encodes semantic content using a pretrained backbone and the other extracts forensic residuals, fused via a lightweight residual adapter for classification, while a shallow Feature Pyramid Network style head produces tamper heatmaps under weak supervision. Red-team training applies worst-of-K counter-forensics per batch, including JPEG realign and recompress, resampling warps, denoise-to-regrain operations, seam smoothing, small color and gamma shifts, and social-app transcodes, while test-time defense injects low-cost jitters such as resize and crop phase changes, mild gamma variation, and JPEG phase shifts with aggregated predictions. Heatmaps are guided to concentrate within face regions using face-box masks without strict pixel-level annotations. Evaluation on existing benchmarks, including standard deepfake datasets and a surveillance-style split with low light and heavy compression, reports clean and attacked performance, AUC, worst-case accuracy, reliability, abstention quality, and weak-localization scores. Results demonstrate near-perfect ranking across attacks, low calibration error, minimal abstention risk, and controlled degradation under regrain, establishing a modular, data-efficient, and practically deployable baseline for attack-aware detection with calibrated probabilities and actionable heatmaps.
Authors: Darrin Bright, Rakshith Raj, Kanchan Keisham
Abstract: Accurate food nutrition estimation from single images is challenging due to the loss of 3D information. While depth-based methods provide reliable geometry, they remain inaccessible on most smartphones because of depth-sensor requirements. To overcome this challenge, we propose PortionNet, a novel cross-modal knowledge distillation framework that learns geometric features from point clouds during training while requiring only RGB images at inference. Our approach employs a dual-mode training strategy where a lightweight adapter network mimics point cloud representations, enabling pseudo-3D reasoning without any specialized hardware requirements. PortionNet achieves state-of-the-art performance on MetaFood3D, outperforming all previous methods in both volume and energy estimation. Cross-dataset evaluation on SimpleFood45 further demonstrates strong generalization in energy estimation.
Authors: Run Ling, Ke Cao, Jian Lu, Ao Ma, Haowei Liu, Runze He, Changwei Wang, Rongtao Xu, Yihua Shao, Zhanjie Zhang, Peng Wu, Guibing Guo, Wei Feng, Zheng Zhang, Jingjing Lv, Junjie Shen, Ching Law, Xingwei Wang
Abstract: Multi-subject video generation aims to synthesize videos from textual prompts and multiple reference images, ensuring that each subject preserves natural scale and visual fidelity. However, current methods face two challenges: scale inconsistency, where variations in subject size lead to unnatural generation, and permutation sensitivity, where the order of reference inputs causes subject distortion. In this paper, we propose MoFu, a unified framework that tackles both challenges. For scale inconsistency, we introduce Scale-Aware Modulation (SMO), an LLM-guided module that extracts implicit scale cues from the prompt and modulates features to ensure consistent subject sizes. To address permutation sensitivity, we present a simple yet effective Fourier Fusion strategy that processes the frequency information of reference features via the Fast Fourier Transform to produce a unified representation. Besides, we design a Scale-Permutation Stability Loss to jointly encourage scale-consistent and permutation-invariant generation. To further evaluate these challenges, we establish a dedicated benchmark with controlled variations in subject scale and reference permutation. Extensive experiments demonstrate that MoFu significantly outperforms existing methods in preserving natural scale, subject fidelity, and overall visual quality.
Authors: Yang Ding, Yizhen Zhang, Xin Lai, Ruihang Chu, Yujiu Yang
Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable progress in vision-language tasks yet remain limited in long video understanding due to the limited context window. Consequently, prevailing approaches tend to rely on uniform frame sampling or static pre-selection, which might overlook critical evidence and unable to correct its initial selection error during its reasoning process. To overcome these limitations, we propose VideoZoomer, a novel agentic framework that enables MLLMs to dynamically control their visual focus during reasoning. Starting from a coarse low-frame-rate overview, VideoZoomer invokes a temporal zoom tool to obtain high-frame-rate clips at autonomously chosen moments, thereby progressively gathering fine-grained evidence in a multi-turn interactive manner. Accordingly, we adopt a two-stage training strategy: a cold-start supervised fine-tuning phase on a curated dataset of distilled exemplar and reflection trajectories, followed by reinforcement learning to further refine the agentic policy. Extensive experiments demonstrate that our 7B model delivers diverse and complex reasoning patterns, yielding strong performance across a broad set of long video understanding and reasoning benchmarks. These emergent capabilities allow it to consistently surpass existing open-source models and even rival proprietary systems on challenging tasks, while achieving superior efficiency under reduced frame budgets.
Authors: Zhibin Qin, Zhenxiong Tan, Zeqing Wang, Songhua Liu, Xinchao Wang
Abstract: Diffusion Transformer models have significantly advanced image editing by encoding conditional images and integrating them into transformer layers. However, most edits involve modifying only small regions, while current methods uniformly process and denoise all tokens at every timestep, causing redundant computation and potentially degrading unchanged areas. This raises a fundamental question: Is it truly necessary to regenerate every region during editing? To address this, we propose SpotEdit, a training-free diffusion editing framework that selectively updates only the modified regions. SpotEdit comprises two key components: SpotSelector identifies stable regions via perceptual similarity and skips their computation by reusing conditional image features; SpotFusion adaptively blends these features with edited tokens through a dynamic fusion mechanism, preserving contextual coherence and editing quality. By reducing unnecessary computation and maintaining high fidelity in unmodified areas, SpotEdit achieves efficient and precise image editing.
Authors: Jianrong Zhang, Hehe Fan, Yi Yang
Abstract: Human motions are compositional: complex behaviors can be described as combinations of simpler primitives. However, existing approaches primarily focus on forward modeling, e.g., learning holistic mappings from text to motion or composing a complex motion from a set of motion concepts. In this paper, we consider the inverse perspective: decomposing a holistic motion into semantically meaningful sub-components. We propose DeMoGen, a compositional training paradigm for decompositional learning that employs an energy-based diffusion model. This energy formulation directly captures the composed distribution of multiple motion concepts, enabling the model to discover them without relying on ground-truth motions for individual concepts. Within this paradigm, we introduce three training variants to encourage a decompositional understanding of motion: 1. DeMoGen-Exp explicitly trains on decomposed text prompts; 2. DeMoGen-OSS performs orthogonal self-supervised decomposition; 3. DeMoGen-SC enforces semantic consistency between original and decomposed text embeddings. These variants enable our approach to disentangle reusable motion primitives from complex motion sequences. We also demonstrate that the decomposed motion concepts can be flexibly recombined to generate diverse and novel motions, generalizing beyond the training distribution. Additionally, we construct a text-decomposed dataset to support compositional training, serving as an extended resource to facilitate text-to-motion generation and motion composition.
Authors: Mariya Miteva, Maria Nisheva-Pavlova
Abstract: Non-invasive inference of molecular tumor characteristics from medical imaging is a central goal of radiogenomics, particularly in glioblastoma (GBM), where O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation carries important prognostic and therapeutic significance. Although radiomics-based machine learning methods have shown promise for this task, conventional unimodal and early-fusion approaches are often limited by high feature redundancy and an incomplete modeling of modality-specific information. In this work, we introduce a multi-view latent representation learning framework based on variational autoencoders (VAE) to integrate complementary radiomic features derived from post-contrast T1-weighted (T1Gd) and Fluid-Attenuated Inversion Recovery (FLAIR) magnetic resonance imaging (MRI). By encoding each modality through an independent probabilistic encoder and performing fusion in a compact latent space, the proposed approach preserves modality-specific structure while enabling effective multimodal integration. The resulting latent embeddings are subsequently used for MGMT promoter methylation classification.
Authors: Olaide N. Oyelade, Oliver Hoxey, Yulia Humrye
Abstract: The popular use of histopathology images, such as hematoxylin and eosin (H&E), has proven to be useful in detecting tumors. However, moving such cancer cases forward for treatment requires accurate on the amount of the human epidermal growth factor receptor 2 (HER2) protein expression. Predicting both the lower and higher levels of HER2 can be challenging. Moreover, jointly analyzing H&E and immunohistochemistry (IHC) stained images for HER2 scoring is difficult. Although several deep learning methods have been investigated to address the challenge of HER2 scoring, they suffer from providing a pixel-level localization of HER2 status. In this study, we propose a single end-to-end pipeline using a system of vision transformers with HER2 status scoring on whole slide images of WSIs. The method includes patch-wise processing of H&E WSIs for tumor localization. A novel mapping function is proposed to correspondingly identify correlated IHC WSIs regions with malignant regions on H&E. A clinically inspired HER2 scoring mechanism is embedded in the pipeline and allows for automatic pixel-level annotation of 4-way HER2 scoring (0, 1+, 2+, and 3+). Also, the proposed method accurately returns HER2-negative and HER2-positive. Privately curated datasets were collaboratively extracted from 13 different cases of WSIs of H&E and IHC. A thorough experiment is conducted on the proposed method. Results obtained showed a good classification accuracy during tumor localization. Also, a classification accuracy of 0.94 and a specificity of 0.933 were returned for the prediction of HER2 status, scoring in the 4-way methods. The applicability of the proposed pipeline was investigated using WSIs patches as comparable to human pathologists. Findings from the study showed the usability of jointly evaluated H&E and IHC images on end-to-end ViTs-based models for HER2 scoring
Authors: Alaa Alahmadi, Mohamed Hasan
Abstract: Machine vision models, particularly deep neural networks, are increasingly applied to physiological signal interpretation, including electrocardiography (ECG), yet they typically require large training datasets and offer limited insight into the causal features underlying their predictions. This lack of data efficiency and interpretability constrains their clinical reliability and alignment with human reasoning. Here, we show that a perception-informed pseudo-colouring technique, previously demonstrated to enhance human ECG interpretation, can improve both explainability and few-shot learning in deep neural networks analysing complex physiological data. We focus on acquired, drug-induced long QT syndrome (LQTS) as a challenging case study characterised by heterogeneous signal morphology, variable heart rate, and scarce positive cases associated with life-threatening arrhythmias such as torsades de pointes. This setting provides a stringent test of model generalisation under extreme data scarcity. By encoding clinically salient temporal features, such as QT-interval duration, into structured colour representations, models learn discriminative and interpretable features from as few as one or five training examples. Using prototypical networks and a ResNet-18 architecture, we evaluate one-shot and few-shot learning on ECG images derived from single cardiac cycles and full 10-second rhythms. Explainability analyses show that pseudo-colouring guides attention toward clinically meaningful ECG features while suppressing irrelevant signal components. Aggregating multiple cardiac cycles further improves performance, mirroring human perceptual averaging across heartbeats. Together, these findings demonstrate that human-like perceptual encoding can bridge data efficiency, explainability, and causal reasoning in medical machine intelligence.
Authors: Zhengfei Kuang, Rui Lin, Long Zhao, Gordon Wetzstein, Saining Xie, Sanghyun Woo
Abstract: Despite the remarkable progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, their application to complex 3D scene manipulation remains underexplored. In this paper, we bridge this critical gap by tackling three key challenges in 3D object arrangement task using MLLMs. First, to address the weak visual grounding of MLLMs, which struggle to link programmatic edits with precise 3D outcomes, we introduce an MCP-based API. This shifts the interaction from brittle raw code manipulation to more robust, function-level updates. Second, we augment the MLLM's 3D scene understanding with a suite of specialized visual tools to analyze scene state, gather spatial information, and validate action outcomes. This perceptual feedback loop is critical for closing the gap between language-based updates and precise 3D-aware manipulation. Third, to manage the iterative, error-prone updates, we propose a collaborative multi-agent framework with designated roles for planning, execution, and verification. This decomposition allows the system to robustly handle multi-step instructions and recover from intermediate errors. We demonstrate the effectiveness of our approach on a diverse set of 25 complex object arrangement tasks, where it significantly outperforms existing baselines. Website: vulcan-3d.github.io
Authors: Xin Yu, Xiaojuan Qi, Zhengqi Li, Kai Zhang, Richard Zhang, Zhe Lin, Eli Shechtman, Tianyu Wang, Yotam Nitzan
Abstract: We introduce the Self-Evaluating Model (Self-E), a novel, from-scratch training approach for text-to-image generation that supports any-step inference. Self-E learns from data similarly to a Flow Matching model, while simultaneously employing a novel self-evaluation mechanism: it evaluates its own generated samples using its current score estimates, effectively serving as a dynamic self-teacher. Unlike traditional diffusion or flow models, it does not rely solely on local supervision, which typically necessitates many inference steps. Unlike distillation-based approaches, it does not require a pretrained teacher. This combination of instantaneous local learning and self-driven global matching bridges the gap between the two paradigms, enabling the training of a high-quality text-to-image model from scratch that excels even at very low step counts. Extensive experiments on large-scale text-to-image benchmarks show that Self-E not only excels in few-step generation, but is also competitive with state-of-the-art Flow Matching models at 50 steps. We further find that its performance improves monotonically as inference steps increase, enabling both ultra-fast few-step generation and high-quality long-trajectory sampling within a single unified model. To our knowledge, Self-E is the first from-scratch, any-step text-to-image model, offering a unified framework for efficient and scalable generation.
Authors: Himanshu Naidu, Yuxiang Zhang, Sachin Mehta, Anat Caspi
Abstract: Accurate, up-to-date sidewalk data is essential for building accessible and inclusive pedestrian infrastructure, yet current approaches to data collection are often costly, fragmented, and difficult to scale. We introduce iOSPointMapper, a mobile application that enables real-time, privacy-conscious sidewalk mapping on the ground, using recent-generation iPhones and iPads. The system leverages on-device semantic segmentation, LiDAR-based depth estimation, and fused GPS/IMU data to detect and localize sidewalk-relevant features such as traffic signs, traffic lights and poles. To ensure transparency and improve data quality, iOSPointMapper incorporates a user-guided annotation interface for validating system outputs before submission. Collected data is anonymized and transmitted to the Transportation Data Exchange Initiative (TDEI), where it integrates seamlessly with broader multimodal transportation datasets. Detailed evaluations of the system's feature detection and spatial mapping performance reveal the application's potential for enhanced pedestrian mapping. Together, these capabilities offer a scalable and user-centered approach to closing critical data gaps in pedestrian
Authors: Hansang Lee, Chaelin Lee, Nieun Seo, Joon Seok Lim, Helen Hong
Abstract: We propose DeFloMat (Detection with Flow Matching), a novel generative object detection framework that addresses the critical latency bottleneck of diffusion-based detectors, such as DiffusionDet, by integrating Conditional Flow Matching (CFM). Diffusion models achieve high accuracy by formulating detection as a multi-step stochastic denoising process, but their reliance on numerous sampling steps ($T \gg 60$) makes them impractical for time-sensitive clinical applications like Crohn's Disease detection in Magnetic Resonance Enterography (MRE). DeFloMat replaces this slow stochastic path with a highly direct, deterministic flow field derived from Conditional Optimal Transport (OT) theory, specifically approximating the Rectified Flow. This shift enables fast inference via a simple Ordinary Differential Equation (ODE) solver. We demonstrate the superiority of DeFloMat on a challenging MRE clinical dataset. Crucially, DeFloMat achieves state-of-the-art accuracy ($43.32\% \text{ } AP_{10:50}$) in only $3$ inference steps, which represents a $1.4\times$ performance improvement over DiffusionDet's maximum converged performance ($31.03\% \text{ } AP_{10:50}$ at $4$ steps). Furthermore, our deterministic flow significantly enhances localization characteristics, yielding superior Recall and stability in the few-step regime. DeFloMat resolves the trade-off between generative accuracy and clinical efficiency, setting a new standard for stable and rapid object localization.
Authors: Amil Khan, Matheus Palhares Viana, Suraj Mishra, B. S. Manjunath
Abstract: Label-free 3D brightfield microscopy offers a fast and noninvasive way to visualize cellular morphology, yet robust volumetric segmentation still typically depends on fluorescence or heavy post-processing. We address this gap by introducing Bright-4B, a 4 billion parameter foundation model that learns on the unit hypersphere to segment subcellular structures directly from 3D brightfield volumes. Bright-4B combines a hardware-aligned Native Sparse Attention mechanism (capturing local, coarse, and selected global context), depth-width residual HyperConnections that stabilize representation flow, and a soft Mixture-of-Experts for adaptive capacity. A plug-and-play anisotropic patch embed further respects confocal point-spread and axial thinning, enabling geometry-faithful 3D tokenization. The resulting model produces morphology-accurate segmentations of nuclei, mitochondria, and other organelles from brightfield stacks alone--without fluorescence, auxiliary channels, or handcrafted post-processing. Across multiple confocal datasets, Bright-4B preserves fine structural detail across depth and cell types, outperforming contemporary CNN and Transformer baselines. All code, pretrained weights, and models for downstream finetuning will be released to advance large-scale, label-free 3D cell mapping.
Authors: Ujunwa Mgboh, Rafi Ibn Sultan, Joshua Kim, Kundan Thind, Dongxiao Zhu
Abstract: Fluence map prediction is central to automated radiotherapy planning but remains an ill-posed inverse problem due to the complex relationship between volumetric anatomy and beam-intensity modulation. Convolutional methods in prior work often struggle to capture long-range dependencies, which can lead to structurally inconsistent or physically unrealizable plans. We introduce \textbf{FluenceFormer}, a backbone-agnostic transformer framework for direct, geometry-aware fluence regression. The model uses a unified two-stage design: Stage~1 predicts a global dose prior from anatomical inputs, and Stage~2 conditions this prior on explicit beam geometry to regress physically calibrated fluence maps. Central to the approach is the \textbf{Fluence-Aware Regression (FAR)} loss, a physics-informed objective that integrates voxel-level fidelity, gradient smoothness, structural consistency, and beam-wise energy conservation. We evaluate the generality of the framework across multiple transformer backbones, including Swin UNETR, UNETR, nnFormer, and MedFormer, using a prostate IMRT dataset. FluenceFormer with Swin UNETR achieves the strongest performance among the evaluated models and improves over existing benchmark CNN and single-stage methods, reducing Energy Error to $\mathbf{4.5\%}$ and yielding statistically significant gains in structural fidelity ($p < 0.05$).
Authors: Jingyuan Yang, Weibin Luo, Hui Huang
Abstract: An image conveys meaning through both its visual content and emotional tone, jointly shaping human perception. We introduce Controllable Emotional Image Content Generation (C-EICG), which aims to generate images that remain faithful to a given content description while expressing a target emotion. Existing text-to-image models ensure content consistency but lack emotional awareness, whereas emotion-driven models generate affective results at the cost of content distortion. To address this gap, we propose EmoCtrl, supported by a dataset annotated with content, emotion, and affective prompts, bridging abstract emotions to visual cues. EmoCtrl incorporates textual and visual emotion enhancement modules that enrich affective expression via descriptive semantics and perceptual cues. The learned emotion tokens exhibit complementary effects, as demonstrated through ablations and visualizations. Quantatitive and qualatitive experiments demonstrate that EmoCtrl achieves faithful content and expressive emotion control, outperforming existing methods across multiple aspects. User studies confirm EmoCtrl's strong alignment with human preference. Moreover, EmoCtrl generalizes well to creative applications, further demonstrating the robustness and adaptability of the learned emotion tokens.
Authors: Khalfalla Awedat, Mohamed Abidalrekab, Gurcan Comert, Mustafa Ayad
Abstract: LiDAR-based perception in autonomous systems is constrained by fixed vertical beam resolution and further compromised by beam dropout resulting from environmental occlusions. This paper introduces SuperiorGAT, a graph attention-based framework designed to reconstruct missing elevation information in sparse LiDAR point clouds. By modeling LiDAR scans as beam-aware graphs and incorporating gated residual fusion with feed-forward refinement, SuperiorGAT enables accurate reconstruction without increasing network depth. To evaluate performance, structured beam dropout is simulated by removing every fourth vertical scanning beam. Extensive experiments across diverse KITTI environments, including Person, Road, Campus, and City sequences, demonstrate that SuperiorGAT consistently achieves lower reconstruction error and improved geometric consistency compared to PointNet-based models and deeper GAT baselines. Qualitative X-Z projections further confirm the model's ability to preserve structural integrity with minimal vertical distortion. These results suggest that architectural refinement offers a computationally efficient method for improving LiDAR resolution without requiring additional sensor hardware.
Authors: Zibin Liu, Banglei Guana, Yang Shanga, Zhenbao Yu, Yifei Bian, Qifeng Yu
Abstract: Camera calibration is an essential prerequisite for event-based vision applications. Current event camera calibration methods typically involve using flashing patterns, reconstructing intensity images, and utilizing the features extracted from events. Existing methods are generally time-consuming and require manually placed calibration objects, which cannot meet the needs of rapidly changing scenarios. In this paper, we propose a line-based event camera calibration framework exploiting the geometric lines of commonly-encountered objects in man-made environments, e.g., doors, windows, boxes, etc. Different from previous methods, our method detects lines directly from event streams and leverages an event-line calibration model to generate the initial guess of camera parameters, which is suitable for both planar and non-planar lines. Then, a non-linear optimization is adopted to refine camera parameters. Both simulation and real-world experiments have demonstrated the feasibility and accuracy of our method, with validation performed on monocular and stereo event cameras. The source code is released at https://github.com/Zibin6/line_based_event_camera_calib.
URLs: https://github.com/Zibin6/line_based_event_camera_calib.
Authors: Zhicheng Zhao, Yuancheng Xu, Andong Lu, Chenglong Li, Jin Tang
Abstract: Optical and Synthetic Aperture Radar (SAR) fusion-based object detection has attracted significant research interest in remote sensing, as these modalities provide complementary information for all-weather monitoring. However, practical deployment is severely limited by inherent challenges. Due to distinct imaging mechanisms, temporal asynchrony, and registration difficulties, obtaining well-aligned optical-SAR image pairs remains extremely difficult, frequently resulting in missing or degraded modality data. Although recent approaches have attempted to address this issue, they still suffer from limited robustness to random missing modalities and lack effective mechanisms to ensure consistent performance improvement in fusion-based detection. To address these limitations, we propose a novel Quality-Aware Dynamic Fusion Network (QDFNet) for robust optical-SAR object detection. Our proposed method leverages learnable reference tokens to dynamically assess feature reliability and guide adaptive fusion in the presence of missing modalities. In particular, we design a Dynamic Modality Quality Assessment (DMQA) module that employs learnable reference tokens to iteratively refine feature reliability assessment, enabling precise identification of degraded regions and providing quality guidance for subsequent fusion. Moreover, we develop an Orthogonal Constraint Normalization Fusion (OCNF) module that employs orthogonal constraints to preserve modality independence while dynamically adjusting fusion weights based on reliability scores, effectively suppressing unreliable feature propagation. Extensive experiments on the SpaceNet6-OTD and OGSOD-2.0 datasets demonstrate the superiority and effectiveness of QDFNet compared to state-of-the-art methods, particularly under partial modality corruption or missing data scenarios.
Authors: Md Abu Obaida Zishan, Annajiat Alim Rasel
Abstract: Locating objects for the visually impaired is a significant challenge and is something no one can get used to over time. However, this hinders their independence and could push them towards risky and dangerous scenarios. Hence, in the spirit of making the visually challenged more self-sufficient, we present SonoVision, a smart-phone application that helps them find everyday objects using sound cues through earphones/headphones. This simply means, if an object is on the right or left side of a user, the app makes a sinusoidal sound in a user's respective ear through ear/headphones. However, to indicate objects located directly in front, both the left and right earphones are rung simultaneously. These sound cues could easily help a visually impaired individual locate objects with the help of their smartphones and reduce the reliance on people in their surroundings, consequently making them more independent. This application is made with the flutter development platform and uses the Efficientdet-D2 model for object detection in the backend. We believe the app will significantly assist the visually impaired in a safe and user-friendly manner with its capacity to work completely offline. Our application can be accessed here https://github.com/MohammedZ666/SonoVision.git.
Authors: Junsheng Yao, Lichao Mou, Qingyu Li
Abstract: Monocular 3D building reconstruction from remote sensing imagery is essential for scalable urban modeling, yet existing methods often require task-specific architectures and intensive supervision. This paper presents the first systematic evaluation of SAM 3D, a general-purpose image-to-3D foundation model, for monocular remote sensing building reconstruction. We benchmark SAM 3D against TRELLIS on samples from the NYC Urban Dataset, employing Frechet Inception Distance (FID) and CLIP-based Maximum Mean Discrepancy (CMMD) as evaluation metrics. Experimental results demonstrate that SAM 3D produces more coherent roof geometry and sharper boundaries compared to TRELLIS. We further extend SAM 3D to urban scene reconstruction through a segment-reconstruct-compose pipeline, demonstrating its potential for urban scene modeling. We also analyze practical limitations and discuss future research directions. These findings provide practical guidance for deploying foundation models in urban 3D reconstruction and motivate future integration of scene-level structural priors.
Authors: Haley Mody (George Mason University), Namish Bansal (George Mason University), Dennies Kiprono Bor (George Mason University), Edward J. Oughton (George Mason University)
Abstract: Electrical substations are a significant component of an electrical grid. Indeed, the assets at these substations (e.g., transformers) are prone to disruption from many hazards, including hurricanes, flooding, earthquakes, and geomagnetically induced currents (GICs). As electrical grids are considered critical national infrastructure, any failure can have significant economic and public safety implications. To help prevent and mitigate these failures, it is thus essential that we identify key substation components to quantify vulnerability. Unfortunately, traditional manual mapping of substation infrastructure is time-consuming and labor-intensive. Therefore, an autonomous solution utilizing computer vision models is preferable, as it allows for greater convenience and efficiency. In this research paper, we train and compare the outputs of 3 models (YOLOv8, YOLOv11, RF-DETR) on a manually labeled dataset of US substation images. Each model is evaluated for detection accuracy, precision, and efficiency. We present the key strengths and limitations of each model, identifying which provides reliable and large-scale substation component mapping. Additionally, we utilize these models to effectively map the various substation components in the United States, showcasing a use case for machine learning in substation mapping.
Authors: Sukhyun Jeong, Yong-Hoon Choi
Abstract: Text-based 3D motion generation aims to automatically synthesize diverse motions from natural-language descriptions to extend user creativity, whereas motion editing modifies an existing motion sequence in response to text while preserving its overall structure. Pose-code-based frameworks such as CoMo map quantifiable pose attributes into discrete pose codes that support interpretable motion control, but their frame-wise representation struggles to capture subtle temporal dynamics and high-frequency details, often degrading reconstruction fidelity and local controllability. To address this limitation, we introduce pose-guided residual refinement for motion (PGR$^2$M), a hybrid representation that augments interpretable pose codes with residual codes learned via residual vector quantization (RVQ). A pose-guided RVQ tokenizer decomposes motion into pose latents that encode coarse global structure and residual latents that model fine-grained temporal variations. Residual dropout further discourages over-reliance on residuals, preserving the semantic alignment and editability of the pose codes. On top of this tokenizer, a base Transformer autoregressively predicts pose codes from text, and a refine Transformer predicts residual codes conditioned on text, pose codes, and quantization stage. Experiments on HumanML3D and KIT-ML show that PGR$^2$M improves Fr\'echet inception distance and reconstruction metrics for both generation and editing compared with CoMo and recent diffusion- and tokenization-based baselines, while user studies confirm that it enables intuitive, structure-preserving motion edits.
Authors: Taihang Lei, Banglei Guan, Minzu Liang, Pengju Sun, Jing Tao, Yang Shang, Qifeng Yu
Abstract: Accurate measurement of shock wave motion parameters with high spatiotemporal resolution is essential for applications such as power field testing and damage assessment. However, significant challenges are posed by the fast, uneven propagation of shock waves and unstable testing conditions. To address these challenges, a novel framework is proposed that utilizes multiple event cameras to estimate the asymmetry of shock waves, leveraging its high-speed and high-dynamic range capabilities. Initially, a polar coordinate system is established, which encodes events to reveal shock wave propagation patterns, with adaptive region-of-interest (ROI) extraction through event offset calculations. Subsequently, shock wave front events are extracted using iterative slope analysis, exploiting the continuity of velocity changes. Finally, the geometric model of events and shock wave motion parameters is derived according to event-based optical imaging model, along with the 3D reconstruction model. Through the above process, multi-angle shock wave measurement, motion field reconstruction, and explosive equivalence inversion are achieved. The results of the speed measurement are compared with those of the pressure sensors and the empirical formula, revealing a maximum error of 5.20% and a minimum error of 0.06%. The experimental results demonstrate that our method achieves high-precision measurement of the shock wave motion field with both high spatial and temporal resolution, representing significant progress.
Authors: Zihan Liu, Xiangning Ren, Dezhang Kong, Yipeng Zhang, Meng Han
Abstract: Infrared small object detection urgently requires semi-supervised paradigms due to the high cost of annotation. However, existing methods like SAM face significant challenges of domain gaps, inability of encoding physical priors, and inherent architectural complexity. To address this, we designed a Hierarchical MoE Adapter consisting of four white-box neural operators. Building upon this core component, we propose a two-stage paradigm for knowledge distillation and transfer: (1) Prior-Guided Knowledge Distillation, where we use our MoE adapter and 10% of available fully supervised data to distill SAM into an expert teacher (Scalpel-SAM); and (2) Deployment-Oriented Knowledge Transfer, where we use Scalpel-SAM to generate pseudo labels for training lightweight and efficient downstream models. Experiments demonstrate that with minimal annotations, our paradigm enables downstream models to achieve performance comparable to, or even surpassing, their fully supervised counterparts. To our knowledge, this is the first semi-supervised paradigm that systematically addresses the data scarcity issue in IR-SOT using SAM as the teacher model.
Authors: Tanish Baranwal, Himanshu Gaurav Singh, Jathushan Rajasegaran, Jitendra Malik
Abstract: We propose Video Gaussian Masked Autoencoders (Video-GMAE), a self-supervised approach for representation learning that encodes a sequence of images into a set of Gaussian splats moving over time. Representing a video as a set of Gaussians enforces a reasonable inductive bias: that 2-D videos are often consistent projections of a dynamic 3-D scene. We find that tracking emerges when pretraining a network with this architecture. Mapping the trajectory of the learnt Gaussians onto the image plane gives zero-shot tracking performance comparable to state-of-the-art. With small-scale finetuning, our models achieve 34.6% improvement on Kinetics, and 13.1% on Kubric datasets, surpassing existing self-supervised video approaches. The project page and code are publicly available at https://videogmae.org/ and https://github.com/tekotan/video-gmae.
URLs: https://videogmae.org/, https://github.com/tekotan/video-gmae.
Authors: Xin Chen, Kang Luo, Yangyi Xiao, Hesheng Wang
Abstract: Reliable and precise detection of small and irregular objects, such as meteor fragments and rocks, is critical for autonomous navigation and operation in lunar surface exploration. Existing multimodal 3D perception methods designed for terrestrial autonomous driving often underperform in off world environments due to poor feature alignment, limited multimodal synergy, and weak small object detection. This paper presents SCAFusion, a multimodal 3D object detection model tailored for lunar robotic missions. Built upon the BEVFusion framework, SCAFusion integrates a Cognitive Adapter for efficient camera backbone tuning, a Contrastive Alignment Module to enhance camera LiDAR feature consistency, a Camera Auxiliary Training Branch to strengthen visual representation, and most importantly, a Section aware Coordinate Attention mechanism explicitly designed to boost the detection performance of small, irregular targets. With negligible increase in parameters and computation, our model achieves 69.7% mAP and 72.1% NDS on the nuScenes validation set, improving the baseline by 5.0% and 2.7%, respectively. In simulated lunar environments built on Isaac Sim, SCAFusion achieves 90.93% mAP, outperforming the baseline by 11.5%, with notable gains in detecting small meteor like obstacles.
Authors: Bin Xia, Bohao Peng, Jiyang Liu, Sitong Wu, Jingyao Li, Junjia Huang, Xu Zhao, Yitong Wang, Ruihang Chu, Bei Yu, Jiaya Jia
Abstract: Recently unified generation and editing models have achieved remarkable success with their impressive performance. These models rely mainly on text prompts for instruction-based editing and generation, but language often fails to capture users intended edit locations and fine-grained visual details. To this end, we propose two tasks: scribble-based editing and generation, that enables more flexible creation on graphical user interface (GUI) combining user textual, images, and freehand sketches. We introduce DreamOmni3, tackling two challenges: data creation and framework design. Our data synthesis pipeline includes two parts: scribble-based editing and generation. For scribble-based editing, we define four tasks: scribble and instruction-based editing, scribble and multimodal instruction-based editing, image fusion, and doodle editing. Based on DreamOmni2 dataset, we extract editable regions and overlay hand-drawn boxes, circles, doodles or cropped image to construct training data. For scribble-based generation, we define three tasks: scribble and instruction-based generation, scribble and multimodal instruction-based generation, and doodle generation, following similar data creation pipelines. For the framework, instead of using binary masks, which struggle with complex edits involving multiple scribbles, images, and instructions, we propose a joint input scheme that feeds both the original and scribbled source images into the model, using different colors to distinguish regions and simplify processing. By applying the same index and position encodings to both images, the model can precisely localize scribbled regions while maintaining accurate editing. Finally, we establish comprehensive benchmarks for these tasks to promote further research. Experimental results demonstrate that DreamOmni3 achieves outstanding performance, and models and code will be publicly released.
Authors: Qinglin Zeng, Kaitong Cai, Ruiqi Chen, Qinhan Lv, Keze Wang
Abstract: Maintaining narrative coherence and visual consistency remains a central challenge in open-domain video generation. Existing text-to-video models often treat each shot independently, resulting in identity drift, scene inconsistency, and unstable temporal structure. We propose CoAgent, a collaborative and closed-loop framework for coherent video generation that formulates the process as a plan-synthesize-verify pipeline. Given a user prompt, style reference, and pacing constraints, a Storyboard Planner decomposes the input into structured shot-level plans with explicit entities, spatial relations, and temporal cues. A Global Context Manager maintains entity-level memory to preserve appearance and identity consistency across shots. Each shot is then generated by a Synthesis Module under the guidance of a Visual Consistency Controller, while a Verifier Agent evaluates intermediate results using vision-language reasoning and triggers selective regeneration when inconsistencies are detected. Finally, a pacing-aware editor refines temporal rhythm and transitions to match the desired narrative flow. Extensive experiments demonstrate that CoAgent significantly improves coherence, visual consistency, and narrative quality in long-form video generation.
Authors: Jesen Zhang, Ningyuan Liu, Kaitong Cai, Sidi Liu, Jing Yang, Ziliang Chen, Xiaofei Sun, Keze Wang
Abstract: Multimodal LLMs often produce fluent yet unreliable reasoning, exhibiting weak step-to-step coherence and insufficient visual grounding, largely because existing alignment approaches supervise only the final answer while ignoring the reliability of the intermediate reasoning process. We introduce SR-MCR, a lightweight and label-free framework that aligns reasoning by exploiting intrinsic process signals derived directly from model outputs. Five self-referential cues -- semantic alignment, lexical fidelity, non-redundancy, visual grounding, and step consistency -- are integrated into a normalized, reliability-weighted reward that provides fine-grained process-level guidance. A critic-free GRPO objective, enhanced with a confidence-aware cooling mechanism, further stabilizes training and suppresses trivial or overly confident generations. Built on Qwen2.5-VL, SR-MCR improves both answer accuracy and reasoning coherence across a broad set of visual benchmarks; among open-source models of comparable size, SR-MCR-7B achieves state-of-the-art performance with an average accuracy of 81.4%. Ablation studies confirm the independent contributions of each reward term and the cooling module.
Authors: Md. Abdur Rahman, Mohaimenul Azam Khan Raiaan, Arefin Ittesafun Abian, Yan Zhang, Mirjam Jonkman, Sami Azam
Abstract: Gliomas are among the most aggressive cancers, characterized by high mortality rates and complex diagnostic processes. Existing studies on glioma diagnosis and classification often describe issues such as high variability in imaging data, inadequate optimization of computational resources, and inefficient segmentation and classification of gliomas. To address these challenges, we propose novel techniques utilizing multi-parametric MRI data to enhance tumor segmentation and classification efficiency. Our work introduces the first-ever radiomics-enhanced fused residual multiparametric 3D network (ReFRM3D) for brain tumor characterization, which is based on a 3D U-Net architecture and features multi-scale feature fusion, hybrid upsampling, and an extended residual skip mechanism. Additionally, we propose a multi-feature tumor marker-based classifier that leverages radiomic features extracted from the segmented regions. Experimental results demonstrate significant improvements in segmentation performance across the BraTS2019, BraTS2020, and BraTS2021 datasets, achieving high Dice Similarity Coefficients (DSC) of 94.04%, 92.68%, and 93.64% for whole tumor (WT), enhancing tumor (ET), and tumor core (TC) respectively in BraTS2019; 94.09%, 92.91%, and 93.84% in BraTS2020; and 93.70%, 90.36%, and 92.13% in BraTS2021.
Authors: Marwan Taher, Ignacio Alzugaray, Kirill Mazur, Xin Kong, Andrew J. Davison
Abstract: Multi-view 3D geometry networks offer a powerful prior but are prohibitively slow for real-time applications. We propose a novel way to adapt them for online use, enabling real-time 6-DoF pose tracking and online reconstruction of objects and scenes from monocular RGB videos. Our method rapidly selects and manages a set of images as keyframes to map a scene or object via $\pi^3$ with full bidirectional attention. We then cache the global self-attention block's key-value (KV) pairs and use them as the sole scene representation for online tracking. This allows for up to $15\times$ speedup during inference without the fear of drift or catastrophic forgetting. Our caching strategy is model-agnostic and can be applied to other off-the-shelf multi-view networks without retraining. We demonstrate KV-Tracker on both scene-level tracking and the more challenging task of on-the-fly object tracking and reconstruction without depth measurements or object priors. Experiments on the TUM RGB-D, 7-Scenes, Arctic and OnePose datasets show the strong performance of our system while maintaining high frame-rates up to ${\sim}27$ FPS.
Authors: Bin Wang, Yang Xu, Huan Zhao, Hao Zhang, Zixing Zhang
Abstract: Speech-driven 3D talking head generation aims to produce lifelike facial animations precisely synchronized with speech. While considerable progress has been made in achieving high lip-synchronization accuracy, existing methods largely overlook the intricate nuances of individual speaking styles, which limits personalization and realism. In this work, we present a novel framework for personalized 3D talking head animation, namely "PTalker". This framework preserves speaking style through style disentanglement from audio and facial motion sequences and enhances lip-synchronization accuracy through a three-level alignment mechanism between audio and mesh modalities. Specifically, to effectively disentangle style and content, we design disentanglement constraints that encode driven audio and motion sequences into distinct style and content spaces to enhance speaking style representation. To improve lip-synchronization accuracy, we adopt a modality alignment mechanism incorporating three aspects: spatial alignment using Graph Attention Networks to capture vertex connectivity in the 3D mesh structure, temporal alignment using cross-attention to capture and synchronize temporal dependencies, and feature alignment by top-k bidirectional contrastive losses and KL divergence constraints to ensure consistency between speech and mesh modalities. Extensive qualitative and quantitative experiments on public datasets demonstrate that PTalker effectively generates realistic, stylized 3D talking heads that accurately match identity-specific speaking styles, outperforming state-of-the-art methods. The source code and supplementary videos are available at: PTalker.
Authors: Dafeng Zhang, Yongqi Song, Shizhuo Liu
Abstract: The method used to measure relationships between face embeddings plays a crucial role in determining the performance of face clustering. Existing methods employ the Jaccard similarity coefficient instead of the cosine distance to enhance the measurement accuracy. However, these methods introduce too many irrelevant nodes, producing Jaccard coefficients with limited discriminative power and adversely affecting clustering performance. To address this issue, we propose a prediction-driven Top-K Jaccard similarity coefficient that enhances the purity of neighboring nodes, thereby improving the reliability of similarity measurements. Nevertheless, accurately predicting the optimal number of neighbors (Top-K) remains challenging, leading to suboptimal clustering results. To overcome this limitation, we develop a Transformer-based prediction model that examines the relationships between the central node and its neighboring nodes near the Top-K to further enhance the reliability of similarity estimation. However, vanilla Transformer, when applied to predict relationships between nodes, often introduces noise due to their overemphasis on irrelevant feature relationships. To address these challenges, we propose a Sparse Differential Transformer (SDT), instead of the vanilla Transformer, to eliminate noise and enhance the model's anti-noise capabilities. Extensive experiments on multiple datasets, such as MS-Celeb-1M, demonstrate that our approach achieves state-of-the-art (SOTA) performance, outperforming existing methods and providing a more robust solution for face clustering.
Authors: Jiacheng Ye, Shansan Gong, Jiahui Gao, Junming Fan, Shuang Wu, Wei Bi, Haoli Bai, Lifeng Shang, Lingpeng Kong
Abstract: While autoregressive Large Vision-Language Models (VLMs) have achieved remarkable success, their sequential generation often limits their efficacy in complex visual planning and dynamic robotic control. In this work, we investigate the potential of constructing Vision-Language Models upon diffusion-based large language models (dLLMs) to overcome these limitations. We introduce Dream-VL, an open diffusion-based VLM (dVLM) that achieves state-of-the-art performance among previous dVLMs. Dream-VL is comparable to top-tier AR-based VLMs trained on open data on various benchmarks but exhibits superior potential when applied to visual planning tasks. Building upon Dream-VL, we introduce Dream-VLA, a dLLM-based Vision-Language-Action model (dVLA) developed through continuous pre-training on open robotic datasets. We demonstrate that the natively bidirectional nature of this diffusion backbone serves as a superior foundation for VLA tasks, inherently suited for action chunking and parallel generation, leading to significantly faster convergence in downstream fine-tuning. Dream-VLA achieves top-tier performance of 97.2% average success rate on LIBERO, 71.4% overall average on SimplerEnv-Bridge, and 60.5% overall average on SimplerEnv-Fractal, surpassing leading models such as $\pi_0$ and GR00T-N1. We also validate that dVLMs surpass AR baselines on downstream tasks across different training objectives. We release both Dream-VL and Dream-VLA to facilitate further research in the community.
Authors: Mohamad Alansari, Muzammal Naseer, Hasan Al Marzouqi, Naoufel Werghi, Sajid Javed
Abstract: \noindent Memory has become the central mechanism enabling robust visual object tracking in modern segmentation-based frameworks. Recent methods built upon Segment Anything Model 2 (SAM2) have demonstrated strong performance by refining how past observations are stored and reused. However, existing approaches address memory limitations in a method-specific manner, leaving the broader design principles of memory in SAM-based tracking poorly understood. Moreover, it remains unclear how these memory mechanisms transfer to stronger, next-generation foundation models such as Segment Anything Model 3 (SAM3). In this work, we present a systematic memory-centric study of SAM-based visual object tracking. We first analyze representative SAM2-based trackers and show that most methods primarily differ in how short-term memory frames are selected, while sharing a common object-centric representation. Building on this insight, we faithfully reimplement these memory mechanisms within the SAM3 framework and conduct large-scale evaluations across ten diverse benchmarks, enabling a controlled analysis of memory design independent of backbone strength. Guided by our empirical findings, we propose a unified hybrid memory framework that explicitly decomposes memory into short-term appearance memory and long-term distractor-resolving memory. This decomposition enables the integration of existing memory policies in a modular and principled manner. Extensive experiments demonstrate that the proposed framework consistently improves robustness under long-term occlusion, complex motion, and distractor-heavy scenarios on both SAM2 and SAM3 backbones. Code is available at: https://github.com/HamadYA/SAM3_Tracking_Zoo. \textbf{This is a preprint. Some results are being finalized and may be updated in a future revision.}
Authors: Yuming Gu, Yizhi Wang, Yining Hong, Yipeng Gao, Hao Jiang, Angtian Wang, Bo Liu, Nathaniel S. Dennler, Zhengfei Kuang, Hao Li, Gordon Wetzstein, Chongyang Ma
Abstract: Embodied visual planning aims to enable manipulation tasks by imagining how a scene evolves toward a desired goal and using the imagined trajectories to guide actions. Video diffusion models, through their image-to-video generation capability, provide a promising foundation for such visual imagination. However, existing approaches are largely forward predictive, generating trajectories conditioned on the initial observation without explicit goal modeling, thus often leading to spatial drift and goal misalignment. To address these challenges, we propose Envision, a diffusion-based framework that performs visual planning for embodied agents. By explicitly constraining the generation with a goal image, our method enforces physical plausibility and goal consistency throughout the generated trajectory. Specifically, Envision operates in two stages. First, a Goal Imagery Model identifies task-relevant regions, performs region-aware cross attention between the scene and the instruction, and synthesizes a coherent goal image that captures the desired outcome. Then, an Env-Goal Video Model, built upon a first-and-last-frame-conditioned video diffusion model (FL2V), interpolates between the initial observation and the goal image, producing smooth and physically plausible video trajectories that connect the start and goal states. Experiments on object manipulation and image editing benchmarks demonstrate that Envision achieves superior goal alignment, spatial consistency, and object preservation compared to baselines. The resulting visual plans can directly support downstream robotic planning and control, providing reliable guidance for embodied agents.
Authors: Yidi Liu, Zihao Fan, Jie Huang, Jie Xiao, Dong Li, Wenlong Zhang, Lei Bai, Xueyang Fu, Zheng-Jun Zha
Abstract: Reinforcement Learning with Human Feedback (RLHF) has proven effective in image generation field guided by reward models to align human preferences. Motivated by this, adapting RLHF for Image Super-Resolution (ISR) tasks has shown promise in optimizing perceptual quality with Image Quality Assessment (IQA) model as reward models. However, the traditional IQA model usually output a single global score, which are exceptionally insensitive to local and fine-grained distortions. This insensitivity allows ISR models to produce perceptually undesirable artifacts that yield spurious high scores, misaligning optimization objectives with perceptual quality and results in reward hacking. To address this, we propose a Fine-grained Perceptual Reward Model (FinPercep-RM) based on an Encoder-Decoder architecture. While providing a global quality score, it also generates a Perceptual Degradation Map that spatially localizes and quantifies local defects. We specifically introduce the FGR-30k dataset to train this model, consisting of diverse and subtle distortions from real-world super-resolution models. Despite the success of the FinPercep-RM model, its complexity introduces significant challenges in generator policy learning, leading to training instability. To address this, we propose a Co-evolutionary Curriculum Learning (CCL) mechanism, where both the reward model and the ISR model undergo synchronized curricula. The reward model progressively increases in complexity, while the ISR model starts with a simpler global reward for rapid convergence, gradually transitioning to the more complex model outputs. This easy-to-hard strategy enables stable training while suppressing reward hacking. Experiments validates the effectiveness of our method across ISR models in both global quality and local realism on RLHF methods.
Authors: Amir El-Ghoussani, Andr\'e Kaup, Nassir Navab, Gustavo Carneiro, Vasileios Belagiannis
Abstract: We propose a monocular depth estimation method based on visual autoregressive (VAR) priors, offering an alternative to diffusion-based approaches. Our method adapts a large-scale text-to-image VAR model and introduces a scale-wise conditional upsampling mechanism with classifier-free guidance. Our approach performs inference in ten fixed autoregressive stages, requiring only 74K synthetic samples for fine-tuning, and achieves competitive results. We report state-of-the-art performance in indoor benchmarks under constrained training conditions, and strong performance when applied to outdoor datasets. This work establishes autoregressive priors as a complementary family of geometry-aware generative models for depth estimation, highlighting advantages in data scalability, and adaptability to 3D vision tasks. Code available at "https://github.com/AmirMaEl/VAR-Depth".
Authors: Shravan Saranyan, Pramit Saha
Abstract: Left ventricular ejection fraction (LVEF) is a key indicator of cardiac function and plays a central role in the diagnosis and management of cardiovascular disease. Echocardiography, as a readily accessible and non-invasive imaging modality, is widely used in clinical practice to estimate LVEF. However, manual assessment of cardiac function from echocardiograms is time-consuming and subject to considerable inter-observer variability. Deep learning approaches offer a promising alternative, with the potential to achieve performance comparable to that of experienced human experts. In this study, we investigate the effectiveness of several deep learning architectures for LVEF estimation from echocardiography videos, including 3D Inception, two-stream, and CNN-RNN models. We systematically evaluate architectural modifications and fusion strategies to identify configurations that maximize prediction accuracy. Models were trained and evaluated on the EchoNet-Dynamic dataset, comprising 10,030 echocardiogram videos. Our results demonstrate that modified 3D Inception architectures achieve the best overall performance, with a root mean squared error (RMSE) of 6.79%. Across architectures, we observe a tendency toward overfitting, with smaller and simpler models generally exhibiting improved generalization. Model performance was also found to be highly sensitive to hyperparameter choices, particularly convolutional kernel sizes and normalization strategies. While this study focuses on echocardiography-based LVEF estimation, the insights gained regarding architectural design and training strategies may be applicable to a broader range of medical and non-medical video analysis tasks.
Authors: Qiankun Li, Feng He, Huabao Chen, Xin Ning, Kun Wang, Zengfu Wang
Abstract: In the big data era, the computer vision field benefits from large-scale datasets such as LAION-2B, LAION-400M, and ImageNet-21K, Kinetics, on which popular models like the ViT and ConvNeXt series have been pre-trained, acquiring substantial knowledge. However, numerous downstream tasks in specialized and data-limited scientific domains continue to pose significant challenges. In this paper, we propose a novel Cluster Attention Adapter (CLAdapter), which refines and adapts the rich representations learned from large-scale data to various data-limited downstream tasks. Specifically, CLAdapter introduces attention mechanisms and cluster centers to personalize the enhancement of transformed features through distribution correlation and transformation matrices. This enables models fine-tuned with CLAdapter to learn distinct representations tailored to different feature sets, facilitating the models' adaptation from rich pre-trained features to various downstream scenarios effectively. In addition, CLAdapter's unified interface design allows for seamless integration with multiple model architectures, including CNNs and Transformers, in both 2D and 3D contexts. Through extensive experiments on 10 datasets spanning domains such as generic, multimedia, biological, medical, industrial, agricultural, environmental, geographical, materials science, out-of-distribution (OOD), and 3D analysis, CLAdapter achieves state-of-the-art performance across diverse data-limited scientific domains, demonstrating its effectiveness in unleashing the potential of foundation vision models via adaptive transfer. Code is available at https://github.com/qklee-lz/CLAdapter.
Authors: Mert Ikinci, Luna Toma, Karin U. Loeffler, Leticia Ussem, Daniela S\"usskind, Julia M. Weller, Yousef Yeganeh, Martina C. Herwig-Carl, Shadi Albarqouni
Abstract: Accurate grading of Conjunctival Melanocytic Intraepithelial Lesions (CMIL) is essential for treatment and melanoma prediction but remains difficult due to subtle morphological cues and interrelated diagnostic criteria. We introduce INTERACT-CMIL, a multi-head deep learning framework that jointly predicts five histopathological axes; WHO4, WHO5, horizontal spread, vertical spread, and cytologic atypia, through Shared Feature Learning with Combinatorial Partial Supervision and an Inter-Dependence Loss enforcing cross-task consistency. Trained and evaluated on a newly curated, multi-center dataset of 486 expert-annotated conjunctival biopsy patches from three university hospitals, INTERACT-CMIL achieves consistent improvements over CNN and foundation-model (FM) baselines, with relative macro F1 gains up to 55.1% (WHO4) and 25.0% (vertical spread). The framework provides coherent, interpretable multi-criteria predictions aligned with expert grading, offering a reproducible computational benchmark for CMIL diagnosis and a step toward standardized digital ocular pathology.
Authors: ZhenQi Chen, TsaiChing Ni, YuanFu Yang
Abstract: Recent text-to-image diffusion models have achieved remarkable visual fidelity but often struggle with semantic alignment to complex prompts. We introduce CritiFusion, a novel inference-time framework that integrates a multimodal semantic critique mechanism with frequency-domain refinement to improve text-to-image consistency and detail. The proposed CritiCore module leverages a vision-language model and multiple large language models to enrich the prompt context and produce high-level semantic feedback, guiding the diffusion process to better align generated content with the prompt's intent. Additionally, SpecFusion merges intermediate generation states in the spectral domain, injecting coarse structural information while preserving high-frequency details. No additional model training is required. CritiFusion serves as a plug-in refinement stage compatible with existing diffusion backbones. Experiments on standard benchmarks show that our method notably improves human-aligned metrics of text-to-image correspondence and visual quality. CritiFusion consistently boosts performance on human preference scores and aesthetic evaluations, achieving results on par with state-of-the-art reward optimization approaches. Qualitative results further demonstrate superior detail, realism, and prompt fidelity, indicating the effectiveness of our semantic critique and spectral alignment strategy.
Authors: Johnathan Xie, Stefan Stojanov, Cristobal Eyzaguirre, Daniel L. K. Yamins, Jiajun Wu
Abstract: Motion prediction has been studied in different contexts with models trained on narrow distributions and applied to downstream tasks in human motion prediction and robotics. Simultaneously, recent efforts in scaling video prediction have demonstrated impressive visual realism, yet they struggle to accurately model complex motions despite massive scale. Inspired by the scaling of video generation, we develop autoregressive flow matching (ARFM), a new method for probabilistic modeling of sequential continuous data and train it on diverse video datasets to generate future point track locations over long horizons. To evaluate our model, we develop benchmarks for evaluating the ability of motion prediction models to predict human and robot motion. Our model is able to predict complex motions, and we demonstrate that conditioning robot action prediction and human motion prediction on predicted future tracks can significantly improve downstream task performance. Code and models publicly available at: https://github.com/Johnathan-Xie/arfm-motion-prediction.
URLs: https://github.com/Johnathan-Xie/arfm-motion-prediction.
Authors: Salvador Rodriguez-Sanz, Monica Hernandez
Abstract: This work proposes a multimodal diffeomorphic registration method using Neural Ordinary Differential Equations (Neural ODEs). Nonrigid registration algorithms exhibit tradeoffs between their accuracy, the computational complexity of their deformation model, and its proper regularization. In addition, they also assume intensity correlation in anatomically homologous regions of interest among image pairs, limiting their applicability to the monomodal setting. Unlike learning-based models, we propose an instance-specific framework that is not subject to high scan requirements for training and does not suffer performance degradation at inference time on modalities unseen during training. Our method exploits the potential of continuous-depth networks in the Neural ODE paradigm with structural descriptors, widely adopted as modality-agnostic metric models which exploit self-similarities on parameterized neighborhood geometries. We propose three different variants that integrate image-based or feature-based structural descriptors and nonstructural image similarities computed by local mutual information. We conduct extensive evaluations on different experiments formed by scan dataset combinations and show surpassing qualitative and quantitative results compared to state-of-the-art baselines adequate for large or small deformations, and specific of multimodal registration. Lastly, we also demonstrate the underlying robustness of the proposed framework to varying levels of explicit regularization while maintaining low error, its suitability for registration at varying scales, and its efficiency with respect to other methods targeted to large-deformation registration.
Authors: Paul Dobre, Jackson Cooper, Xin Wang, Hongzhou Yang
Abstract: 3D Asset insertion and novel view synthesis (NVS) are key components for autonomous driving simulation, enhancing the diversity of training data. With better training data that is diverse and covers a wide range of situations, including long-tailed driving scenarios, autonomous driving models can become more robust and safer. This motivates a unified simulation framework that can jointly handle realistic integration of inserted 3D assets and NVS. Recent 3D asset reconstruction methods enable reconstruction of dynamic actors from video, supporting their re-insertion into simulated driving scenes. While the overall structure and appearance can be accurate, it still struggles to capture the realism of 3D assets through lighting or shadows, particularly when inserted into scenes. In parallel, recent advances in NVS methods have demonstrated promising results in synthesizing viewpoints beyond the originally recorded trajectories. However, existing approaches largely treat asset insertion and NVS capabilities in isolation. To allow for interaction with the rest of the scene and to enable more diverse creation of new scenarios for training, realistic 3D asset insertion should be combined with NVS. To address this, we present SCPainter (Street Car Painter), a unified framework which integrates 3D Gaussian Splat (GS) car asset representations and 3D scene point clouds with diffusion-based generation to jointly enable realistic 3D asset insertion and NVS. The 3D GS assets and 3D scene point clouds are projected together into novel views, and these projections are used to condition a diffusion model to generate high quality images. Evaluation on the Waymo Open Dataset demonstrate the capability of our framework to enable 3D asset insertion and NVS, facilitating the creation of diverse and realistic driving data.
Authors: Youssef Megahed, Robin Ducharme, Inok Lee, Inbal Willner, Olivier X. Miguel, Kevin Dick, Adrian D. C. Chan, Mark Walker, Steven Hawken
Abstract: Cystic hygroma is a high-risk prenatal ultrasound finding that portends high rates of chromosomal abnormalities, structural malformations, and adverse pregnancy outcomes. Automated detection can increase reproducibility and support scalable early screening programs, but supervised deep learning methods are limited by small labelled datasets. This study assesses whether ultrasound-specific self-supervised pretraining can facilitate accurate, robust deep learning detection of cystic hygroma in first-trimester ultrasound images. We fine-tuned the Ultrasound Self-Supervised Foundation Model with Masked Autoencoding (USF-MAE), pretrained on over 370,000 unlabelled ultrasound images, for binary classification of normal controls and cystic hygroma cases used in this study. Performance was evaluated on the same curated ultrasound dataset, preprocessing pipeline, and 4-fold cross-validation protocol as for the DenseNet-169 baseline, using accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (ROC-AUC). Model interpretability was analyzed qualitatively using Score-CAM visualizations. USF-MAE outperformed the DenseNet-169 baseline on all evaluation metrics. The proposed model yielded a mean accuracy of 0.96, sensitivity of 0.94, specificity of 0.98, and ROC-AUC of 0.98 compared to 0.93, 0.92, 0.94, and 0.94 for the DenseNet-169 baseline, respectively. Qualitative Score-CAM visualizations of model predictions demonstrated clinical relevance by highlighting expected regions in the fetal neck for both positive and negative cases. Paired statistical analysis using a Wilcoxon signed-rank test confirmed that performance improvements achieved by USF-MAE were statistically significant (p = 0.0057).
Authors: Yongzhen Hu, Yihui Yang, Haotong Lin, Yifan Wang, Junting Dong, Yifu Deng, Xinyu Zhu, Fan Jia, Hujun Bao, Xiaowei Zhou, Sida Peng
Abstract: This paper addresses the problem of decomposed 4D scene reconstruction from multi-view videos. Recent methods achieve this by lifting video segmentation results to a 4D representation through differentiable rendering techniques. Therefore, they heavily rely on the quality of video segmentation maps, which are often unstable, leading to unreliable reconstruction results. To overcome this challenge, our key idea is to represent the decomposed 4D scene with the Freetime FeatureGS and design a streaming feature learning strategy to accurately recover it from per-image segmentation maps, eliminating the need for video segmentation. Freetime FeatureGS models the dynamic scene as a set of Gaussian primitives with learnable features and linear motion ability, allowing them to move to neighboring regions over time. We apply a contrastive loss to Freetime FeatureGS, forcing primitive features to be close or far apart based on whether their projections belong to the same instance in the 2D segmentation map. As our Gaussian primitives can move across time, it naturally extends the feature learning to the temporal dimension, achieving 4D segmentation. Furthermore, we sample observations for training in a temporally ordered manner, enabling the streaming propagation of features over time and effectively avoiding local minima during the optimization process. Experimental results on several datasets show that the reconstruction quality of our method outperforms recent methods by a large margin.
Authors: Hao Zhang, Mengsi Lyu, Bo Huang, Yulong Ao, Yonghua Lin
Abstract: Large Multimodal Models (LMMs) have proven effective on various tasks. They typically encode visual inputs into Original Model sequences of tokens, which are then concatenated with textual tokens and jointly processed by the language model. However, the growing number of visual tokens greatly increases inference cost. Visual token pruning has emerged as a promising solution. However, existing methods often overlook scenarios involving long context inputs with multiple images. In this paper, we analyze the challenges of visual token pruning in long context, multi-image settings and introduce an adaptive pruning method tailored for such scenarios. We decompose redundancy into intra-image and inter-image components and quantify them through intra-image diversity and inter-image variation, which jointly guide dynamic budget allocation. Our approach consists of two stages. The intra-image stage allocates each image a content-aware token budget and greedily selects its most representative tokens. The inter-image stage performs global diversity filtering to form a candidate pool and then applies a Pareto selection procedure that balances diversity with text alignment. Extensive experiments show that our approach maintains strong performance in long context settings while significantly cutting down the number of visual tokens.
Authors: Yunge Li, Lanyu Xu
Abstract: Vision Transformers (ViTs) have achieved remarkable success in visual recognition tasks, but redundant token representations limit their computational efficiency. Existing token merging and pruning strategies often overlook spatial continuity and neighbor relationships, resulting in the loss of local context. This paper proposes novel neighbor-aware token reduction methods based on Hilbert curve reordering, which explicitly preserves the neighbor structure in a 2D space using 1D sequential representations. Our method introduces two key strategies: Neighbor-Aware Pruning (NAP) for selective token retention and Merging by Adjacent Token similarity (MAT) for local token aggregation. Experiments demonstrate that our approach achieves state-of-the-art accuracy-efficiency trade-offs compared to existing methods. This work highlights the importance of spatial continuity and neighbor structure, offering new insights for the architectural optimization of ViTs.
Authors: Yiqian Li, Wen Jiang, Kostas Daniilidis
Abstract: Understanding semantics and dynamics has been crucial for embodied agents in various tasks. Both tasks have much more data redundancy than the static scene understanding task. We formulate the view selection problem as an active learning problem, where the goal is to prioritize frames that provide the greatest information gain for model training. To this end, we propose an active learning algorithm with Fisher Information that quantifies the informativeness of candidate views with respect to both semantic Gaussian parameters and deformation networks. This formulation allows our method to jointly handle semantic reasoning and dynamic scene modeling, providing a principled alternative to heuristic or random strategies. We evaluate our method on large-scale static images and dynamic video datasets by selecting informative frames from multi-camera setups. Experimental results demonstrate that our approach consistently improves rendering quality and semantic segmentation performance, outperforming baseline methods based on random selection and uncertainty-based heuristics.
Authors: Zhicheng Liao, Baoliang Chen, Hanwei Zhu, Lingyu Zhu, Shiqi Wang, Weisi Lin
Abstract: Existing AGIQA models typically estimate image quality by measuring and aggregating the similarities between image embeddings and text embeddings derived from multi-grade quality descriptions. Although effective, we observe that such similarity distributions across grades usually exhibit multimodal patterns. For instance, an image embedding may show high similarity to both "excellent" and "poor" grade descriptions while deviating from the "good" one. We refer to this phenomenon as "semantic drift", where semantic inconsistencies between text embeddings and their intended descriptions undermine the reliability of text-image shared-space learning. To mitigate this issue, we draw inspiration from psychometrics and propose an improved Graded Response Model (GRM) for AGIQA. The GRM is a classical assessment model that categorizes a subject's ability across grades using test items with various difficulty levels. This paradigm aligns remarkably well with human quality rating, where image quality can be interpreted as an image's ability to meet various quality grades. Building on this philosophy, we design a two-branch quality grading module: one branch estimates image ability while the other constructs multiple difficulty levels. To ensure monotonicity in difficulty levels, we further model difficulty generation in an arithmetic manner, which inherently enforces a unimodal and interpretable quality distribution. Our Arithmetic GRM based Quality Grading (AGQG) module enjoys a plug-and-play advantage, consistently improving performance when integrated into various state-of-the-art AGIQA frameworks. Moreover, it also generalizes effectively to both natural and screen content image quality assessment, revealing its potential as a key component in future IQA models.
Authors: Ruoyu Wang, Ziyu Li, Beier Zhu, Liangyu Yuan, Hanwang Zhang, Xun Yang, Xiaojun Chang, Chi Zhang
Abstract: Diffusion models (DMs) have achieved state-of-the-art generative performance but suffer from high sampling latency due to their sequential denoising nature. Existing solver-based acceleration methods often face significant image quality degradation under a low-latency budget, primarily due to accumulated truncation errors arising from the inability to capture high-curvature trajectory segments. In this paper, we propose the Ensemble Parallel Direction solver (dubbed as EPD-Solver), a novel ODE solver that mitigates these errors by incorporating multiple parallel gradient evaluations in each step. Motivated by the geometric insight that sampling trajectories are largely confined to a low-dimensional manifold, EPD-Solver leverages the Mean Value Theorem for vector-valued functions to approximate the integral solution more accurately. Importantly, since the additional gradient computations are independent, they can be fully parallelized, preserving low-latency sampling nature. We introduce a two-stage optimization framework. Initially, EPD-Solver optimizes a small set of learnable parameters via a distillation-based approach. We further propose a parameter-efficient Reinforcement Learning (RL) fine-tuning scheme that reformulates the solver as a stochastic Dirichlet policy. Unlike traditional methods that fine-tune the massive backbone, our RL approach operates strictly within the low-dimensional solver space, effectively mitigating reward hacking while enhancing performance in complex text-to-image (T2I) generation tasks. In addition, our method is flexible and can serve as a plugin (EPD-Plugin) to improve existing ODE samplers.
Authors: Jingchao Wang, Kaiwen Zhou, Zhijian Wu, Kunhua Ji, Dingjiang Huang, Yefeng Zheng
Abstract: Vision-Language Tracking aims to continuously localize objects described by a visual template and a language description. Existing methods, however, are typically limited to local search, making them prone to failures under viewpoint changes, occlusions, and rapid target movements. In this work, we introduce the first global tracking framework based on Multimodal Large Language Models (VPTracker), exploiting their powerful semantic reasoning to locate targets across the entire image space. While global search improves robustness and reduces drift, it also introduces distractions from visually or semantically similar objects. To address this, we propose a location-aware visual prompting mechanism that incorporates spatial priors into the MLLM. Specifically, we construct a region-level prompt based on the target's previous location, enabling the model to prioritize region-level recognition and resort to global inference only when necessary. This design retains the advantages of global tracking while effectively suppressing interference from distracting visual content. Extensive experiments show that our approach significantly enhances tracking stability and target disambiguation under challenging scenarios, opening a new avenue for integrating MLLMs into visual tracking. Code is available at https://github.com/jcwang0602/VPTracker.
Authors: Bin Liu, Wenyan Tian, Huangxin Fu, Zizheng Li, Zhifen He, Bo Li
Abstract: 3D reconstruction of medical images is a key technology in medical image analysis and clinical diagnosis, providing structural visualization support for disease assessment and surgical planning. Traditional methods are computationally expensive and prone to structural discontinuities and loss of detail in sparse slices, making it difficult to meet clinical accuracy requirements.To address these challenges, we propose an efficient 3D reconstruction method based on 3D Gaussian and tri-plane representations. This method not only maintains the advantages of Gaussian representation in efficient rendering and geometric representation but also significantly enhances structural continuity and semantic consistency under sparse slicing conditions. Experimental results on multimodal medical datasets such as US and MRI show that our proposed method can generate high-quality, anatomically coherent, and semantically stable medical images under sparse data conditions, while significantly improving reconstruction efficiency. This provides an efficient and reliable new approach for 3D visualization and clinical analysis of medical images.
Authors: Po-Chih Wu
Abstract: Open-vocabulary object detection enables models to localize and recognize objects beyond a predefined set of categories and is expected to achieve recognition capabilities comparable to human performance. In this study, we aim to evaluate the performance of existing models on open-vocabulary object detection tasks under low-quality image conditions. For this purpose, we introduce a new dataset that simulates low-quality images in the real world. In our evaluation experiment, we find that although open-vocabulary object detection models exhibited no significant decrease in mAP scores under low-level image degradation, the performance of all models dropped sharply under high-level image degradation. OWLv2 models consistently performed better across different types of degradation, while OWL-ViT, GroundingDINO, and Detic showed significant performance declines. We will release our dataset and codes to facilitate future studies.
Authors: Libo Zhang, Zekun Li, Tianyu Li, Zeyu Cao, Rui Xu, Xiaoxiao Long, Wenjia Wang, Jingbo Wang, Yuan Liu, Wenping Wang, Daquan Zhou, Taku Komura, Zhiyang Dou
Abstract: Humans exhibit adaptive, context-sensitive responses to egocentric visual input. However, faithfully modeling such reactions from egocentric video remains challenging due to the dual requirements of strictly causal generation and precise 3D spatial alignment. To tackle this problem, we first construct the Human Reaction Dataset (HRD) to address data scarcity and misalignment by building a spatially aligned egocentric video-reaction dataset, as existing datasets (e.g., ViMo) suffer from significant spatial inconsistency between the egocentric video and reaction motion, e.g., dynamically moving motions are always paired with fixed-camera videos. Leveraging HRD, we present EgoReAct, the first autoregressive framework that generates 3D-aligned human reaction motions from egocentric video streams in real-time. We first compress the reaction motion into a compact yet expressive latent space via a Vector Quantised-Variational AutoEncoder and then train a Generative Pre-trained Transformer for reaction generation from the visual input. EgoReAct incorporates 3D dynamic features, i.e., metric depth, and head dynamics during the generation, which effectively enhance spatial grounding. Extensive experiments demonstrate that EgoReAct achieves remarkably higher realism, spatial consistency, and generation efficiency compared with prior methods, while maintaining strict causality during generation. We will release code, models, and data upon acceptance.
Authors: Hualie Jiang, Ziyang Song, Zhiqiang Lou, Rui Xu, Minglang Tan
Abstract: Panoramic depth estimation provides a comprehensive solution for capturing complete $360^\circ$ environmental structural information, offering significant benefits for robotics and AR/VR applications. However, while extensively studied in indoor settings, its zero-shot generalization to open-world domains lags far behind perspective images, which benefit from abundant training data. This disparity makes transferring capabilities from the perspective domain an attractive solution. To bridge this gap, we present Depth Anything in $360^\circ$ (DA360), a panoramic-adapted version of Depth Anything V2. Our key innovation involves learning a shift parameter from the ViT backbone, transforming the model's scale- and shift-invariant output into a scale-invariant estimate that directly yields well-formed 3D point clouds. This is complemented by integrating circular padding into the DPT decoder to eliminate seam artifacts, ensuring spatially coherent depth maps that respect spherical continuity. Evaluated on standard indoor benchmarks and our newly curated outdoor dataset, Metropolis, DA360 shows substantial gains over its base model, achieving over 50\% and 10\% relative depth error reduction on indoor and outdoor benchmarks, respectively. Furthermore, DA360 significantly outperforms robust panoramic depth estimation methods, achieving about 30\% relative error improvement compared to PanDA across all three test datasets and establishing new state-of-the-art performance for zero-shot panoramic depth estimation.
Authors: Chenyu Li, Danfeng Hong, Bing Zhang, Zhaojie Pan, Jocelyn Chanussot
Abstract: The highly nonlinear degradation process, complex physical interactions, and various sources of uncertainty render single-image Super-resolution (SR) a particularly challenging task. Existing interpretable SR approaches, whether based on prior learning or deep unfolding optimization frameworks, typically rely on black-box deep networks to model latent variables, which leaves the degradation process largely unknown and uncontrollable. Inspired by the Kolmogorov-Arnold theorem (KAT), we for the first time propose a novel interpretable operator, termed Kolmogorov-Arnold Neural Operator (KANO), with the application to image SR. KANO provides a transparent and structured representation of the latent degradation fitting process. Specifically, we employ an additive structure composed of a finite number of B-spline functions to approximate continuous spectral curves in a piecewise fashion. By learning and optimizing the shape parameters of these spline functions within defined intervals, our KANO accurately captures key spectral characteristics, such as local linear trends and the peak-valley structures at nonlinear inflection points, thereby endowing SR results with physical interpretability. Furthermore, through theoretical modeling and experimental evaluations across natural images, aerial photographs, and satellite remote sensing data, we systematically compare multilayer perceptrons (MLPs) and Kolmogorov-Arnold networks (KANs) in handling complex sequence fitting tasks. This comparative study elucidates the respective advantages and limitations of these models in characterizing intricate degradation mechanisms, offering valuable insights for the development of interpretable SR techniques.
Authors: Zirui Zhou, Junfeng Ni, Shujie Zhang, Yixin Chen, Siyuan Huang
Abstract: Change detection plays a vital role in scene monitoring, exploration, and continual reconstruction. Existing 3D change detection methods often exhibit spatial inconsistency in the detected changes and fail to explicitly separate pre- and post-change states. To address these limitations, we propose SCaR-3D, a novel 3D scene change detection framework that identifies object-level changes from a dense-view pre-change image sequence and sparse-view post-change images. Our approach consists of a signed-distance-based 2D differencing module followed by multi-view aggregation with voting and pruning, leveraging the consistent nature of 3DGS to robustly separate pre- and post-change states. We further develop a continual scene reconstruction strategy that selectively updates dynamic regions while preserving the unchanged areas. We also contribute CCS3D, a challenging synthetic dataset that allows flexible combinations of 3D change types to support controlled evaluations. Extensive experiments demonstrate that our method achieves both high accuracy and efficiency, outperforming existing methods.
Authors: Zhenbao Yu, Shirong Ye, Ronghe Jin, Shunkun Liang, Zibin Liu, Huiyun Zhang, Banglei Guan
Abstract: In this paper, we aim to estimate the relative pose and focal length between two views with known intrinsic parameters except for an unknown focal length from two affine correspondences (ACs). Cameras are commonly used in combination with inertial measurement units (IMUs) in applications such as self-driving cars, smartphones, and unmanned aerial vehicles. The vertical direction of camera views can be obtained by IMU measurements. The relative pose between two cameras is reduced from 5DOF to 3DOF. We propose a new solver to estimate the 3DOF relative pose and focal length. First, we establish constraint equations from two affine correspondences when the vertical direction is known. Then, based on the properties of the equation system with nontrivial solutions, four equations can be derived. These four equations only involve two parameters: the focal length and the relative rotation angle. Finally, the polynomial eigenvalue method is utilized to solve the problem of focal length and relative rotation angle. The proposed solver is evaluated using synthetic and real-world datasets. The results show that our solver performs better than the existing state-of-the-art solvers.
Authors: Bangya Liu, Xinyu Gong, Zelin Zhao, Ziyang Song, Yulei Lu, Suhui Wu, Jun Zhang, Suman Banerjee, Hao Zhang
Abstract: Human-object interaction (HOI) video generation has garnered increasing attention due to its promising applications in digital humans, e-commerce, advertising, and robotics imitation learning. However, existing methods face two critical limitations: (1) a lack of effective mechanisms to inject multi-view information of the object into the model, leading to poor cross-view consistency, and (2) heavy reliance on fine-grained hand mesh annotations for modeling interaction occlusions. To address these challenges, we introduce ByteLoom, a Diffusion Transformer (DiT)-based framework that generates realistic HOI videos with geometrically consistent object illustration, using simplified human conditioning and 3D object inputs. We first propose an RCM-cache mechanism that leverages Relative Coordinate Maps (RCM) as a universal representation to maintain object's geometry consistency and precisely control 6-DoF object transformations in the meantime. To compensate HOI dataset scarcity and leverage existing datasets, we further design a training curriculum that enhances model capabilities in a progressive style and relaxes the demand of hand mesh. Extensive experiments demonstrate that our method faithfully preserves human identity and the object's multi-view geometry, while maintaining smooth motion and object manipulation.
Authors: Zhuonan Liu, Xinyu Zhang, Zishuo Wang, Tomohito Kawabata, Xuesu Xiao, Ling Xiao
Abstract: Socially compliant navigation requires structured reasoning over dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. However, existing social navigation datasets often lack explicit reasoning supervision and exhibit highly long-tailed action distributions, limiting models' ability to learn safety-critical behaviors. To address these issues, we introduce MUSON, a multimodal dataset for short-horizon social navigation collected across diverse indoor and outdoor campus scenes. MUSON adopts a structured five-step Chain-of-Thought annotation consisting of perception, prediction, reasoning, action, and explanation, with explicit modeling of static physical constraints and a rationally balanced discrete action space. Compared to SNEI, MUSON provides consistent reasoning, action, and explanation. Benchmarking multiple state-of-the-art Small Vision Language Models on MUSON shows that Qwen2.5-VL-3B achieves the highest decision accuracy of 0.8625, demonstrating that MUSON serves as an effective and reusable benchmark for socially compliant navigation. The dataset is publicly available at https://huggingface.co/datasets/MARSLab/MUSON
Authors: Ziyu Zhou, Haozhe Luo, Mohammad Reza Hosseinzadeh Taher, Jiaxuan Pang, Xiaowei Ding, Michael B. Gotway, Jianming Liang
Abstract: Foundation models have been successful in natural language processing and computer vision because they are capable of capturing the underlying structures (foundation) of natural languages. However, in medical imaging, the key foundation lies in human anatomy, as these images directly represent the internal structures of the body, reflecting the consistency, coherence, and hierarchy of human anatomy. Yet, existing self-supervised learning (SSL) methods often overlook these perspectives, limiting their ability to effectively learn anatomical features. To overcome the limitation, we built Lamps (learning anatomy from multiple perspectives via self-supervision) pre-trained on large-scale chest radiographs by harmoniously utilizing the consistency, coherence, and hierarchy of human anatomy as the supervision signal. Extensive experiments across 10 datasets evaluated through fine-tuning and emergent property analysis demonstrate Lamps' superior robustness, transferability, and clinical potential when compared to 10 baseline models. By learning from multiple perspectives, Lamps presents a unique opportunity for foundation models to develop meaningful, robust representations that are aligned with the structure of human anatomy.
Authors: Weiwei Li, Junzhuo Liu, Yuanyuan Ren, Yuchen Zheng, Yahao Liu, Wen Li
Abstract: Deep learning models are known to often learn features that spuriously correlate with the class label during training but are irrelevant to the prediction task. Existing methods typically address this issue by annotating potential spurious attributes, or filtering spurious features based on some empirical assumptions (e.g., simplicity of bias). However, these methods may yield unsatisfactory performance due to the intricate and elusive nature of spurious correlations in real-world data. In this paper, we propose a data-oriented approach to mitigate the spurious correlation in deep learning models. We observe that samples that are influenced by spurious features tend to exhibit a dispersed distribution in the learned feature space. This allows us to identify the presence of spurious features. Subsequently, we obtain a bias-invariant representation by neutralizing the spurious features based on a simple grouping strategy. Then, we learn a feature transformation to eliminate the spurious features by aligning with this bias-invariant representation. Finally, we update the classifier by incorporating the learned feature transformation and obtain an unbiased model. By integrating the aforementioned identifying, neutralizing, eliminating and updating procedures, we build an effective pipeline for mitigating spurious correlation. Experiments on image and NLP debiasing benchmarks show an improvement in worst group accuracy of more than 20% compared to standard empirical risk minimization (ERM). Codes and checkpoints are available at https://github.com/davelee-uestc/nsf_debiasing .
Authors: Ju-Hsuan Weng, Jia-Wei Liao, Cheng-Fu Chou, Jun-Cheng Chen
Abstract: Text-to-image diffusion models may generate harmful or copyrighted content, motivating research on concept erasure. However, existing approaches primarily focus on erasing concepts from text prompts, overlooking other input modalities that are increasingly critical in real-world applications such as image editing and personalized generation. These modalities can become attack surfaces, where erased concepts re-emerge despite defenses. To bridge this gap, we introduce M-ErasureBench, a novel multimodal evaluation framework that systematically benchmarks concept erasure methods across three input modalities: text prompts, learned embeddings, and inverted latents. For the latter two, we evaluate both white-box and black-box access, yielding five evaluation scenarios. Our analysis shows that existing methods achieve strong erasure performance against text prompts but largely fail under learned embeddings and inverted latents, with Concept Reproduction Rate (CRR) exceeding 90% in the white-box setting. To address these vulnerabilities, we propose IRECE (Inference-time Robustness Enhancement for Concept Erasure), a plug-and-play module that localizes target concepts via cross-attention and perturbs the associated latents during denoising. Experiments demonstrate that IRECE consistently restores robustness, reducing CRR by up to 40% under the most challenging white-box latent inversion scenario, while preserving visual quality. To the best of our knowledge, M-ErasureBench provides the first comprehensive benchmark of concept erasure beyond text prompts. Together with IRECE, our benchmark offers practical safeguards for building more reliable protective generative models.
Authors: Hasan Faraz Khan, Noor Fatima, Muzammil Behzad
Abstract: The recent integration of artificial intelligence into medical imaging has driven remarkable advances in automated organ segmentation. However, most existing 3D segmentation frameworks rely exclusively on visual learning from large annotated datasets restricting their adaptability to new domains and clinical tasks. The lack of semantic understanding in these models makes them ineffective in addressing flexible, user-defined segmentation objectives. To overcome these limitations, we propose SwinTF3D, a lightweight multimodal fusion approach that unifies visual and linguistic representations for text-guided 3D medical image segmentation. The model employs a transformer-based visual encoder to extract volumetric features and integrates them with a compact text encoder via an efficient fusion mechanism. This design allows the system to understand natural-language prompts and correctly align semantic cues with their corresponding spatial structures in medical volumes, while producing accurate, context-aware segmentation results with low computational overhead. Extensive experiments on the BTCV dataset demonstrate that SwinTF3D achieves competitive Dice and IoU scores across multiple organs, despite its compact architecture. The model generalizes well to unseen data and offers significant efficiency gains compared to conventional transformer-based segmentation networks. Bridging visual perception with linguistic understanding, SwinTF3D establishes a practical and interpretable paradigm for interactive, text-driven 3D medical image segmentation, opening perspectives for more adaptive and resource-efficient solutions in clinical imaging.
Authors: Haosen Li, Wenshuo Chen, Shaofeng Liang, Lei Wang, Haozhe Jia, Yutao Yue
Abstract: Iterative refinement methods based on a denoising-inversion cycle are powerful tools for enhancing the quality and control of diffusion models. However, their effectiveness is critically limited when combined with standard Classifier-Free Guidance (CFG). We identify a fundamental limitation: CFG's extrapolative nature systematically pushes the sampling path off the data manifold, causing the approximation error to diverge and undermining the refinement process. To address this, we propose Guided Path Sampling (GPS), a new paradigm for iterative refinement. GPS replaces unstable extrapolation with a principled, manifold-constrained interpolation, ensuring the sampling path remains on the data manifold. We theoretically prove that this correction transforms the error series from unbounded amplification to strictly bounded, guaranteeing stability. Furthermore, we devise an optimal scheduling strategy that dynamically adjusts guidance strength, aligning semantic injection with the model's natural coarse-to-fine generation process. Extensive experiments on modern backbones like SDXL and Hunyuan-DiT show that GPS outperforms existing methods in both perceptual quality and complex prompt adherence. For instance, GPS achieves a superior ImageReward of 0.79 and HPS v2 of 0.2995 on SDXL, while improving overall semantic alignment accuracy on GenEval to 57.45%. Our work establishes that path stability is a prerequisite for effective iterative refinement, and GPS provides a robust framework to achieve it.
Authors: Yangzhi Ma, Bojun Liu, Jie Li, Li Li, Dong Liu
Abstract: Hash grids are widely used to learn an implicit neural field for Gaussian splatting, serving either as part of the entropy model or for inter-frame prediction. However, due to the irregular and non-uniform distribution of Gaussian splats in 3D space, numerous sparse regions exist, rendering many features in the hash grid invalid. This leads to redundant storage and transmission overhead. In this work, we propose a hash grid feature pruning method that identifies and prunes invalid features based on the coordinates of the input Gaussian splats, so that only the valid features are encoded. This approach reduces the storage size of the hash grid without compromising model performance, leading to improved rate-distortion performance. Following the Common Test Conditions (CTC) defined by the standardization committee, our method achieves an average bitrate reduction of 8% compared to the baseline approach.
Authors: Kai Liu, Jungang Li, Yuchong Sun, Shengqiong Wu, Jianzhang Gao, Daoan Zhang, Wei Zhang, Sheng Jin, Sicheng Yu, Geng Zhan, Jiayi Ji, Fan Zhou, Liang Zheng, Shuicheng Yan, Hao Fei, Tat-Seng Chua
Abstract: This paper presents JavisGPT, the first unified multimodal large language model (MLLM) for Joint Audio-Video (JAV) comprehension and generation. JavisGPT adopts a concise encoder-LLM-decoder architecture, featuring a SyncFusion module for spatio-temporal audio-video fusion and synchrony-aware learnable queries to bridge a pretrained JAV-DiT generator. This design enables temporally coherent video-audio understanding and generation from multimodal instructions. We design an effective three-stage training pipeline consisting of multimodal pretraining, audio-video fine-tuning, and large-scale instruction-tuning, to progressively build multimodal comprehension and generation from existing vision-language models. To support this, we further construct JavisInst-Omni, a high-quality instruction dataset with over 200K GPT-4o-curated audio-video-text dialogues that span diverse and multi-level comprehension and generation scenarios. Extensive experiments on JAV comprehension and generation benchmarks show that JavisGPT outperforms existing MLLMs, particularly in complex and temporally synchronized settings.
Authors: Qihang Peng, Xuesong Chen, Chenye Yang, Shaoshuai Shi, Hongsheng Li
Abstract: Autonomous driving requires generating safe and reliable trajectories from complex multimodal inputs. Traditional modular pipelines separate perception, prediction, and planning, while recent end-to-end (E2E) systems learn them jointly. Vision-language models (VLMs) further enrich this paradigm by introducing cross-modal priors and commonsense reasoning, yet current VLM-based planners face three key challenges: (i) a mismatch between discrete text reasoning and continuous control, (ii) high latency from autoregressive chain-of-thought decoding, and (iii) inefficient or non-causal planners that limit real-time deployment. We propose ColaVLA, a unified vision-language-action framework that transfers reasoning from text to a unified latent space and couples it with a hierarchical, parallel trajectory decoder. The Cognitive Latent Reasoner compresses scene understanding into compact, decision-oriented meta-action embeddings through ego-adaptive selection and only two VLM forward passes. The Hierarchical Parallel Planner then generates multi-scale, causality-consistent trajectories in a single forward pass. Together, these components preserve the generalization and interpretability of VLMs while enabling efficient, accurate and safe trajectory generation. Experiments on the nuScenes benchmark show that ColaVLA achieves state-of-the-art performance in both open-loop and closed-loop settings with favorable efficiency and robustness.
Authors: Zhicheng Zhao, Xuanang Fan, Lingma Sun, Chenglong Li, Jin Tang
Abstract: High-resolution remote sensing imagery increasingly contains dense clusters of tiny objects, the detection of which is extremely challenging due to severe mutual occlusion and limited pixel footprints. Existing detection methods typically allocate computational resources uniformly, failing to adaptively focus on these density-concentrated regions, which hinders feature learning effectiveness. To address these limitations, we propose the Dense Region Mining Network (DRMNet), which leverages density maps as explicit spatial priors to guide adaptive feature learning. First, we design a Density Generation Branch (DGB) to model object distribution patterns, providing quantifiable priors that guide the network toward dense regions. Second, to address the computational bottleneck of global attention, our Dense Area Focusing Module (DAFM) uses these density maps to identify and focus on dense areas, enabling efficient local-global feature interaction. Finally, to mitigate feature degradation during hierarchical extraction, we introduce a Dual Filter Fusion Module (DFFM). It disentangles multi-scale features into high- and low-frequency components using a discrete cosine transform and then performs density-guided cross-attention to enhance complementarity while suppressing background interference. Extensive experiments on the AI-TOD and DTOD datasets demonstrate that DRMNet surpasses state-of-the-art methods, particularly in complex scenarios with high object density and severe occlusion.
Authors: Behnam Raoufi, Hossein Sharify, Mohamad Mahdee Ramezanee, Khosrow Hajsadeghi, Saeed Bagheri Shouraki
Abstract: Conventional object detectors rely on cross-entropy classification, which can be vulnerable to class imbalance and label noise. We propose CLIP-Joint-Detect, a simple and detector-agnostic framework that integrates CLIP-style contrastive vision-language supervision through end-to-end joint training. A lightweight parallel head projects region or grid features into the CLIP embedding space and aligns them with learnable class-specific text embeddings via InfoNCE contrastive loss and an auxiliary cross-entropy term, while all standard detection losses are optimized simultaneously. The approach applies seamlessly to both two-stage and one-stage architectures. We validate it on Pascal VOC 2007+2012 using Faster R-CNN and on the large-scale MS COCO 2017 benchmark using modern YOLO detectors (YOLOv11), achieving consistent and substantial improvements while preserving real-time inference speed. Extensive experiments and ablations demonstrate that joint optimization with learnable text embeddings markedly enhances closed-set detection performance across diverse architectures and datasets.
Authors: Runwei Guan, Jianan Liu, Shaofeng Liang, Fangqiang Ding, Shanliang Yao, Xiaokai Bai, Daizong Liu, Tao Huang, Guoqiang Mao, Hui Xiong
Abstract: 4D millimeter-wave (mmWave) radar has been widely adopted in autonomous driving and robot perception due to its low cost and all-weather robustness. However, its inherent sparsity and limited semantic richness significantly constrain perception capability. Recently, fusing camera data with 4D radar has emerged as a promising cost effective solution, by exploiting the complementary strengths of the two modalities. Nevertheless, point-cloud-based radar often suffer from information loss introduced by multi-stage signal processing, while directly utilizing raw 4D radar data incurs prohibitive computational costs. To address these challenges, we propose WRCFormer, a novel 3D object detection framework that fuses raw radar cubes with camera inputs via multi-view representations of the decoupled radar cube. Specifically, we design a Wavelet Attention Module as the basic module of wavelet-based Feature Pyramid Network (FPN) to enhance the representation of sparse radar signals and image data. We further introduce a two-stage query-based, modality-agnostic fusion mechanism termed Geometry-guided Progressive Fusion to efficiently integrate multi-view features from both modalities. Extensive experiments demonstrate that WRCFormer achieves state-of-the-art performance on the K-Radar benchmarks, surpassing the best model by approximately 2.4% in all scenarios and 1.6% in the sleet scenario, highlighting its robustness under adverse weather conditions.
Authors: Shizhou Zhang, Xueqiang Lv, Yinghui Xing, Qirui Wu, Di Xu, Chen Zhao, Yanning Zhang
Abstract: Current methods for incremental object detection (IOD) primarily rely on Faster R-CNN or DETR series detectors; however, these approaches do not accommodate the real-time YOLO detection frameworks. In this paper, we first identify three primary types of knowledge conflicts that contribute to catastrophic forgetting in YOLO-based incremental detectors: foreground-background confusion, parameter interference, and misaligned knowledge distillation. Subsequently, we introduce YOLO-IOD, a real-time Incremental Object Detection (IOD) framework that is constructed upon the pretrained YOLO-World model, facilitating incremental learning via a stage-wise parameter-efficient fine-tuning process. Specifically, YOLO-IOD encompasses three principal components: 1) Conflict-Aware Pseudo-Label Refinement (CPR), which mitigates the foreground-background confusion by leveraging the confidence levels of pseudo labels and identifying potential objects relevant to future tasks. 2) Importancebased Kernel Selection (IKS), which identifies and updates the pivotal convolution kernels pertinent to the current task during the current learning stage. 3) Cross-Stage Asymmetric Knowledge Distillation (CAKD), which addresses the misaligned knowledge distillation conflict by transmitting the features of the student target detector through the detection heads of both the previous and current teacher detectors, thereby facilitating asymmetric distillation between existing and newly introduced categories. We further introduce LoCo COCO, a more realistic benchmark that eliminates data leakage across stages. Experiments on both conventional and LoCo COCO benchmarks show that YOLO-IOD achieves superior performance with minimal forgetting.
Authors: Chunyuan Chen, Yunuo Cai, Shujuan Li, Weiyun Liang, Bin Wang, Jing Xu
Abstract: Camouflaged image generation (CIG) has recently emerged as an efficient alternative for acquiring high-quality training data for camouflaged object detection (COD). However, existing CIG methods still suffer from a substantial gap to real camouflaged imagery: generated images either lack sufficient camouflage due to weak visual similarity, or exhibit cluttered backgrounds that are semantically inconsistent with foreground targets. To address these limitations, we propose ReamCamo, a unified out-painting based framework for realistic camouflaged image generation. ReamCamo explicitly introduces additional layout controls to regulate global image structure, thereby improving semantic coherence between foreground objects and generated backgrounds. Moreover, we construct a multi-modal textual-visual condition by combining a unified fine-grained textual task description with texture-oriented background retrieval, which jointly guides the generation process to enhance visual fidelity and realism. To quantitatively assess camouflage quality, we further introduce a background-foreground distribution divergence metric that measures the effectiveness of camouflage in generated images. Extensive experiments and visualizations demonstrate the effectiveness of our proposed framework.
Authors: Huiming Yang, Linglin Liao, Fei Ding, Sibo Wang, Zijian Zeng
Abstract: Six degree of freedom (6DoF) pose estimation for novel objects is a critical task in computer vision, yet it faces significant challenges in high-speed and low-light scenarios where standard RGB cameras suffer from motion blur. While event cameras offer a promising solution due to their high temporal resolution, current 6DoF pose estimation methods typically yield suboptimal performance in high-speed object moving scenarios. To address this gap, we propose PoseStreamer, a robust multi-modal 6DoF pose estimation framework designed specifically on high-speed moving scenarios. Our approach integrates three core components: an Adaptive Pose Memory Queue that utilizes historical orientation cues for temporal consistency, an Object-centric 2D Tracker that provides strong 2D priors to boost 3D center recall, and a Ray Pose Filter for geometric refinement along camera rays. Furthermore, we introduce MoCapCube6D, a novel multi-modal dataset constructed to benchmark performance under rapid motion. Extensive experiments demonstrate that PoseStreamer not only achieves superior accuracy in high-speed moving scenarios, but also exhibits strong generalizability as a template-free framework for unseen moving objects.
Authors: Linglin Liao, Qichuan Geng, Yu Liu
Abstract: Text-guided Medical Image Segmentation has shown considerable promise for medical image segmentation, with rich clinical text serving as an effective supplement for scarce data. However, current methods have two key bottlenecks. On one hand, they struggle to process diagnostic and descriptive texts simultaneously, making it difficult to identify lesions and establish associations with image regions. On the other hand, existing approaches focus on lesions description and fail to capture positional constraints, leading to critical deviations. Specifically, with the text "in the left lower lung", the segmentation results may incorrectly cover both sides of the lung. To address the limitations, we propose the Spatial-aware Symmetric Alignment (SSA) framework to enhance the capacity of referring hybrid medical texts consisting of locational, descriptive, and diagnostic information. Specifically, we propose symmetric optimal transport alignment mechanism to strengthen the associations between image regions and multiple relevant expressions, which establishes bi-directional fine-grained multimodal correspondences. In addition, we devise a composite directional guidance strategy that explicitly introduces spatial constraints in the text by constructing region-level guidance masks. Extensive experiments on public benchmarks demonstrate that SSA achieves state-of-the-art (SOTA) performance, particularly in accurately segmenting lesions characterized by spatial relational constraints.
Authors: Han-Wei Kung, Tuomas Varanka, Nicu Sebe
Abstract: Recent text-to-image diffusion models have demonstrated remarkable generation of realistic facial images conditioned on textual prompts and human identities, enabling creating personalized facial imagery. However, existing prompt-based methods for removing or modifying identity-specific features rely either on the subject being well-represented in the pre-trained model or require model fine-tuning for specific identities. In this work, we analyze the identity generation process and introduce a reverse personalization framework for face anonymization. Our approach leverages conditional diffusion inversion, allowing direct manipulation of images without using text prompts. To generalize beyond subjects in the model's training data, we incorporate an identity-guided conditioning branch. Unlike prior anonymization methods, which lack control over facial attributes, our framework supports attribute-controllable anonymization. We demonstrate that our method achieves a state-of-the-art balance between identity removal, attribute preservation, and image quality. Source code and data are available at https://github.com/hanweikung/reverse-personalization .
Authors: Soham Dutta, Soham Banerjee, Sneha Mahata, Anindya Sen, Sayantani Datta
Abstract: Apple orchards require timely disease detection, fruit quality assessment, and yield estimation, yet existing UAV-based systems address such tasks in isolation and often rely on costly multispectral sensors. This paper presents a unified, low-cost RGB-only UAV-based orchard intelligent pipeline integrating ResNet50 for leaf disease detection, VGG 16 for apple freshness determination, and YOLOv8 for real-time apple detection and localization. The system runs on an ESP32-CAM and Raspberry Pi, providing fully offline on-site inference without cloud support. Experiments demonstrate 98.9% accuracy for leaf disease classification, 97.4% accuracy for freshness classification, and 0.857 F1 score for apple detection. The framework provides an accessible and scalable alternative to multispectral UAV solutions, supporting practical precision agriculture on affordable hardware.
Authors: Wenyuan Huang, Zhao Wang, Zhou Wei, Ting Huang, Fang Zhao, Jian Yang, Zhenyu Zhang
Abstract: 3D visual grounding aims to locate objects based on natural language descriptions in 3D scenes. Existing methods rely on a pre-defined Object Lookup Table (OLT) to query Visual Language Models (VLMs) for reasoning about object locations, which limits the applications in scenarios with undefined or unforeseen targets. To address this problem, we present OpenGround, a novel zero-shot framework for open-world 3D visual grounding. Central to OpenGround is the Active Cognition-based Reasoning (ACR) module, which is designed to overcome the fundamental limitation of pre-defined OLTs by progressively augmenting the cognitive scope of VLMs. The ACR module performs human-like perception of the target via a cognitive task chain and actively reasons about contextually relevant objects, thereby extending VLM cognition through a dynamically updated OLT. This allows OpenGround to function with both pre-defined and open-world categories. We also propose a new dataset named OpenTarget, which contains over 7000 object-description pairs to evaluate our method in open-world scenarios. Extensive experiments demonstrate that OpenGround achieves competitive performance on Nr3D, state-of-the-art on ScanRefer, and delivers a substantial 17.6% improvement on OpenTarget. Project Page at [this https URL](https://why-102.github.io/openground.io/).
Authors: Ciprian Constantinescu, Marius Leordeanu
Abstract: Humans effortlessly identify objects by leveraging a rich understanding of the surrounding scene, including spatial relationships, material properties, and the co-occurrence of other objects. In contrast, most computational object recognition systems operate on isolated image regions, devoid of meaning in isolation, thus ignoring this vital contextual information. This paper argues for the critical role of context and introduces a novel framework for contextual object classification. We first construct a Geo-Semantic Contextual Graph (GSCG) from a single monocular image. This rich, structured representation is built by integrating a metric depth estimator with a unified panoptic and material segmentation model. The GSCG encodes objects as nodes with detailed geometric, chromatic, and material attributes, and their spatial relationships as edges. This explicit graph structure makes the model's reasoning process inherently interpretable. We then propose a specialized graph-based classifier that aggregates features from a target object, its immediate neighbors, and the global scene context to predict its class. Through extensive ablation studies, we demonstrate that our context-aware model achieves a classification accuracy of 73.4%, dramatically outperforming context-agnostic versions (as low as 38.4%). Furthermore, our GSCG-based approach significantly surpasses strong baselines, including fine-tuned ResNet models (max 53.5%) and a state-of-the-art multimodal Large Language Model (LLM), Llama 4 Scout, which, even when given the full image alongside a detailed description of objects, maxes out at 42.3%. These results on COCO 2017 train/val splits highlight the superiority of explicitly structured and interpretable context for object recognition tasks.
Authors: Thomson Tong, Diba Darooneh
Abstract: This report provides an architecture-led analysis of two modern vision-language models (VLMs), Qwen2.5-VL-7B-Instruct and Llama-4-Scout-17B-16E-Instruct, and explains how their architectural properties map to a practical video-to-artifact pipeline implemented in the BodyLanguageDetection repository [1]. The system samples video frames, prompts a VLM to detect visible people and generate pixel-space bounding boxes with prompt-conditioned attributes (emotion by default), validates output structure using a predefined schema, and optionally renders an annotated video. We first summarize the shared multimodal foundation (visual tokenization, Transformer attention, and instruction following), then describe each architecture at a level sufficient to justify engineering choices without speculative internals. Finally, we connect model behavior to system constraints: structured outputs can be syntactically valid while semantically incorrect, schema validation is structural (not geometric correctness), person identifiers are frame-local in the current prompting contract, and interactive single-frame analysis returns free-form text rather than schema-enforced JSON. These distinctions are critical for writing defensible claims, designing robust interfaces, and planning evaluation.
Authors: Yi Zhou, Xuechao Zou, Shun Zhang, Kai Li, Shiying Wang, Jingming Chen, Congyan Lang, Tengfei Cao, Pin Tao, Yuanchun Shi
Abstract: Semi-supervised remote sensing (RS) image semantic segmentation offers a promising solution to alleviate the burden of exhaustive annotation, yet it fundamentally struggles with pseudo-label drift, a phenomenon where confirmation bias leads to the accumulation of errors during training. In this work, we propose Co2S, a stable semi-supervised RS segmentation framework that synergistically fuses priors from vision-language models and self-supervised models. Specifically, we construct a heterogeneous dual-student architecture comprising two distinct ViT-based vision foundation models initialized with pretrained CLIP and DINOv3 to mitigate error accumulation and pseudo-label drift. To effectively incorporate these distinct priors, an explicit-implicit semantic co-guidance mechanism is introduced that utilizes text embeddings and learnable queries to provide explicit and implicit class-level guidance, respectively, thereby jointly enhancing semantic consistency. Furthermore, a global-local feature collaborative fusion strategy is developed to effectively fuse the global contextual information captured by CLIP with the local details produced by DINOv3, enabling the model to generate highly precise segmentation results. Extensive experiments on six popular datasets demonstrate the superiority of the proposed method, which consistently achieves leading performance across various partition protocols and diverse scenarios. Project page is available at https://xavierjiezou.github.io/Co2S/.
Authors: Ryousuke Yamada, Kohsuke Ide, Yoshihiro Fukuhara, Hirokatsu Kataoka, Gilles Puy, Andrei Bursuc, Yuki M. Asano
Abstract: Despite recent progress in 3D self-supervised learning, collecting large-scale 3D scene scans remains expensive and labor-intensive. In this work, we investigate whether 3D representations can be learned from unlabeled videos recorded without any real 3D sensors. We present Laplacian-Aware Multi-level 3D Clustering with Sinkhorn-Knopp (LAM3C), a self-supervised framework that learns from video-generated point clouds from unlabeled videos. We first introduce RoomTours, a video-generated point cloud dataset constructed by collecting room-walkthrough videos from the web (e.g., real-estate tours) and generating 49,219 scenes using an off-the-shelf feed-forward reconstruction model. We also propose a noise-regularized loss that stabilizes representation learning by enforcing local geometric smoothness and ensuring feature stability under noisy point clouds. Remarkably, without using any real 3D scans, LAM3C achieves higher performance than the previous self-supervised methods on indoor semantic and instance segmentation. These results suggest that unlabeled videos represent an abundant source of data for 3D self-supervised learning.
Authors: Zhengyang Liang, Yan Shu, Xiangrui Liu, Minghao Qin, Kaixin Liang, Paolo Rota, Nicu Sebe, Zheng Liu, Lizi Liao
Abstract: The evolution of autonomous agents is redefining information seeking, transitioning from passive retrieval to proactive, open-ended web research. However, while textual and static multimodal agents have seen rapid progress, a significant modality gap remains in processing the web's most dynamic modality: video. Existing video benchmarks predominantly focus on passive perception, feeding curated clips to models without requiring external retrieval. They fail to evaluate agentic video research, which necessitates actively interrogating video timelines, cross-referencing dispersed evidence, and verifying claims against the open web. To bridge this gap, we present \textbf{Video-BrowseComp}, a challenging benchmark comprising 210 questions tailored for open-web agentic video reasoning. Unlike prior benchmarks, Video-BrowseComp enforces a mandatory dependency on temporal visual evidence, ensuring that answers cannot be derived solely through text search but require navigating video timelines to verify external claims. Our evaluation of state-of-the-art models reveals a critical bottleneck: even advanced search-augmented models like GPT-5.1 (w/ Search) achieve only 15.24\% accuracy. Our analysis reveals that these models largely rely on textual proxies, excelling in metadata-rich domains (e.g., TV shows with plot summaries) but collapsing in metadata-sparse, dynamic environments (e.g., sports, gameplay) where visual grounding is essential. As the first open-web video research benchmark, Video-BrowseComp advances the field beyond passive perception toward proactive video reasoning.
Authors: Brayden Miao, Zain Rehman, Xin Miao, Siming Liu, Jianjie Wang
Abstract: Chest X-ray (CXR) imaging is widely used for screening and diagnosing pulmonary abnormalities, yet automated interpretation remains challenging due to weak disease signals, dataset bias, and limited spatial supervision. Foundation models for medical image segmentation (MedSAM) provide an opportunity to introduce anatomically grounded priors that may improve robustness and interpretability in CXR analysis. We propose a segmentation-guided CXR classification pipeline that integrates MedSAM as a lung region extraction module prior to multi-label abnormality classification. MedSAM is fine-tuned using a public image-mask dataset from Airlangga University Hospital. We then apply it to a curated subset of the public NIH CXR dataset to train and evaluate deep convolutional neural networks for multi-label prediction of five abnormalities (Mass, Nodule, Pneumonia, Edema, and Fibrosis), with the normal case (No Finding) evaluated via a derived score. Experiments show that MedSAM produces anatomically plausible lung masks across diverse imaging conditions. We find that masking effects are both task-dependent and architecture-dependent. ResNet50 trained on original images achieves the strongest overall abnormality discrimination, while loose lung masking yields comparable macro AUROC but significantly improves No Finding discrimination, indicating a trade-off between abnormality-specific classification and normal case screening. Tight masking consistently reduces abnormality level performance but improves training efficiency. Loose masking partially mitigates this degradation by preserving perihilar and peripheral context. These results suggest that lung masking should be treated as a controllable spatial prior selected to match the backbone and clinical objective, rather than applied uniformly.
Authors: Jian Wang, Sixing Rong, Jiarui Xing, Yuling Xu, Weide Liu
Abstract: We present PathoSyn, a unified generative framework for Magnetic Resonance Imaging (MRI) image synthesis that reformulates imaging-pathology as a disentangled additive deviation on a stable anatomical manifold. Current generative models typically operate in the global pixel domain or rely on binary masks, these paradigms often suffer from feature entanglement, leading to corrupted anatomical substrates or structural discontinuities. PathoSyn addresses these limitations by decomposing the synthesis task into deterministic anatomical reconstruction and stochastic deviation modeling. Central to our framework is a Deviation-Space Diffusion Model designed to learn the conditional distribution of pathological residuals, thereby capturing localized intensity variations while preserving global structural integrity by construction. To ensure spatial coherence, the diffusion process is coupled with a seam-aware fusion strategy and an inference-time stabilization module, which collectively suppress boundary artifacts and produce high-fidelity internal lesion heterogeneity. PathoSyn provides a mathematically principled pipeline for generating high-fidelity patient-specific synthetic datasets, facilitating the development of robust diagnostic algorithms in low-data regimes. By allowing interpretable counterfactual disease progression modeling, the framework supports precision intervention planning and provides a controlled environment for benchmarking clinical decision-support systems. Quantitative and qualitative evaluations on tumor imaging benchmarks demonstrate that PathoSyn significantly outperforms holistic diffusion and mask-conditioned baselines in both perceptual realism and anatomical fidelity. The source code of this work will be made publicly available.
Authors: Mingzhen Shao, Sarang Joshi
Abstract: Deep learning has advanced deformable image registration, surpassing traditional optimization-based methods in both accuracy and efficiency. However, learning-based models are widely believed to be sensitive to domain shift, with robustness typically pursued through large and diverse training datasets, without explaining the underlying mechanisms. In this work, we show that domain-shift immunity is an inherent property of deep deformable registration models, arising from their reliance on local feature representations rather than global appearance for deformation estimation. To isolate and validate this mechanism, we introduce UniReg, a universal registration framework that decouples feature extraction from deformation estimation using fixed, pre-trained feature extractors and a UNet-based deformation network. Despite training on a single dataset, UniReg exhibits robust cross-domain and multi-modal performance comparable to optimization-based methods. Our analysis further reveals that failures of conventional CNN-based models under modality shift originate from dataset-induced biases in early convolutional layers. These findings identify local feature consistency as the key driver of robustness in learning-based deformable registration and motivate backbone designs that preserve domain-invariant local features.
Authors: Jingyu Li, Xiaolong Zhao, Zhe Liu, Wenxiao Wu, Li Zhang
Abstract: Semi-supervised 3D object detection, aiming to explore unlabeled data for boosting 3D object detectors, has emerged as an active research area in recent years. Some previous methods have shown substantial improvements by either employing heterogeneous teacher models to provide high-quality pseudo labels or enforcing feature-perspective consistency between the teacher and student networks. However, these methods overlook the fact that the model usually tends to exhibit low sensitivity to object geometries with limited labeled data, making it difficult to capture geometric information, which is crucial for enhancing the student model's ability in object perception and localization. In this paper, we propose GeoTeacher to enhance the student model's ability to capture geometric relations of objects with limited training data, especially unlabeled data. We design a keypoint-based geometric relation supervision module that transfers the teacher model's knowledge of object geometry to the student, thereby improving the student's capability in understanding geometric relations. Furthermore, we introduce a voxel-wise data augmentation strategy that increases the diversity of object geometries, thereby further improving the student model's ability to comprehend geometric structures. To preserve the integrity of distant objects during augmentation, we incorporate a distance-decay mechanism into this strategy. Moreover, GeoTeacher can be combined with different SS3D methods to further improve their performance. Extensive experiments on the ONCE and Waymo datasets indicate the effectiveness and generalization of our method and we achieve the new state-of-the-art results. Code will be available at https://github.com/SII-Whaleice/GeoTeacher
Authors: Fulin Shi, Wenyi Xiao, Bin Chen, Liang Din, Leilei Gan
Abstract: Evaluating the alignment between textual prompts and generated images is critical for ensuring the reliability and usability of text-to-image (T2I) models. However, most existing evaluation methods rely on coarse-grained metrics or static QA pipelines, which lack fine-grained interpretability and struggle to reflect human preferences. To address this, we propose REVEALER, a unified framework for element-level alignment evaluation based on reinforcement-guided visual reasoning. Adopting a structured "grounding-reasoning-conclusion" paradigm, our method enables Multimodal Large Language Models (MLLMs) to explicitly localize semantic elements and derive interpretable alignment judgments. We optimize the model via Group Relative Policy Optimization(GRPO) using a composite reward function that incorporates structural format, grounding accuracy, and alignment fidelity. Extensive experiments across four benchmarks-EvalMuse-40K, RichHF, MHaluBench, and GenAI-Bench-demonstrate that REVEALER achieves state-of-the-art performance. Our approach consistently outperforms both strong proprietary models and supervised baselines while demonstrating superior inference efficiency compared to existing iterative visual reasoning methods.
Authors: Yi Zhang, Yi Wang, Lei Yao, Lap-Pui Chau
Abstract: Image-based 3D object detection aims to identify and localize objects in 3D space using only RGB images, eliminating the need for expensive depth sensors required by point cloud-based methods. Existing image-based approaches face two critical challenges: methods achieving high accuracy typically require dense 3D supervision, while those operating without such supervision struggle to extract accurate geometry from images alone. In this paper, we present GVSynergy-Det, a novel framework that enhances 3D detection through synergistic Gaussian-Voxel representation learning. Our key insight is that continuous Gaussian and discrete voxel representations capture complementary geometric information: Gaussians excel at modeling fine-grained surface details while voxels provide structured spatial context. We introduce a dual-representation architecture that: 1) adapts generalizable Gaussian Splatting to extract complementary geometric features for detection tasks, and 2) develops a cross-representation enhancement mechanism that enriches voxel features with geometric details from Gaussian fields. Unlike previous methods that either rely on time-consuming per-scene optimization or utilize Gaussian representations solely for depth regularization, our synergistic strategy directly leverages features from both representations through learnable integration, enabling more accurate object localization. Extensive experiments demonstrate that GVSynergy-Det achieves state-of-the-art results on challenging indoor benchmarks, significantly outperforming existing methods on both ScanNetV2 and ARKitScenes datasets, all without requiring any depth or dense 3D geometry supervision (e.g., point clouds or TSDF).
Authors: Tianchen Deng, Xuefeng Chen, Yi Chen, Qu Chen, Yuyao Xu, Lijin Yang, Le Xu, Yu Zhang, Bo Zhang, Wuxiong Huang, Hesheng Wang
Abstract: Driving World Models (DWMs) have been developing rapidly with the advances of generative models. However, existing DWMs lack 3D scene understanding capabilities and can only generate content conditioned on input data, without the ability to interpret or reason about the driving environment. Moreover, current approaches represent 3D spatial information with point cloud or BEV features do not accurately align textual information with the underlying 3D scene. To address these limitations, we propose a novel unified DWM framework based on 3D Gaussian scene representation, which enables both 3D scene understanding and multi-modal scene generation, while also enabling contextual enrichment for understanding and generation tasks. Our approach directly aligns textual information with the 3D scene by embedding rich linguistic features into each Gaussian primitive, thereby achieving early modality alignment. In addition, we design a novel task-aware language-guided sampling strategy that removes redundant 3D Gaussians and injects accurate and compact 3D tokens into LLM. Furthermore, we design a dual-condition multi-modal generation model, where the information captured by our vision-language model is leveraged as a high-level language condition in combination with a low-level image condition, jointly guiding the multi-modal generation process. We conduct comprehensive studies on the nuScenes, and NuInteract datasets to validate the effectiveness of our framework. Our method achieves state-of-the-art performance. We will release the code publicly on GitHub https://github.com/dtc111111/GaussianDWM.
Authors: Maisha Haque, Israt Jahan Ayshi, Sadaf M. Anis, Nahian Tasnim, Mithila Moontaha, Md. Sabbir Ahmed, Muhammad Iqbal Hossain, Mohammad Zavid Parvez, Subrata Chakraborty, Biswajeet Pradhan, Biswajit Banik
Abstract: This research proposes "ForCM", a novel approach to forest cover mapping that combines Object-Based Image Analysis (OBIA) with Deep Learning (DL) using multispectral Sentinel-2 imagery. The study explores several DL models, including UNet, UNet++, ResUNet, AttentionUNet, and ResNet50-Segnet, applied to high-resolution Sentinel-2 Level 2A satellite images of the Amazon Rainforest. The datasets comprise three collections: two sets of three-band imagery and one set of four-band imagery. After evaluation, the most effective DL models are individually integrated with the OBIA technique to enhance mapping accuracy. The originality of this work lies in evaluating different deep learning models combined with OBIA and comparing them with traditional OBIA methods. The results show that the proposed ForCM method improves forest cover mapping, achieving overall accuracies of 94.54 percent with ResUNet-OBIA and 95.64 percent with AttentionUNet-OBIA, compared to 92.91 percent using traditional OBIA. This research also demonstrates the potential of free and user-friendly tools such as QGIS for accurate mapping within their limitations, supporting global environmental monitoring and conservation efforts.
Authors: Jongoh Jeong, Youngjin Oh, Gyeongrae Nam, Jeongeun Lee, Kuk-Jin Yoon
Abstract: Object detection is one of the key target tasks of interest in the context of civil and military applications. In particular, the real-world deployment of target detection methods is pivotal in the decision-making process during military command and reconnaissance. However, current domain adaptive object detection algorithms consider adapting one domain to another similar one only within the scope of natural or autonomous driving scenes. Since military domains often deal with a mixed variety of environments, detecting objects from multiple varying target domains poses a greater challenge. Several studies for armored military target detection have made use of synthetic aperture radar (SAR) data due to its robustness to all weather, long range, and high-resolution characteristics. Nevertheless, the costs of SAR data acquisition and processing are still much higher than those of the conventional RGB camera, which is a more affordable alternative with significantly lower data processing time. Furthermore, the lack of military target detection datasets limits the use of such a low-cost approach. To mitigate these issues, we propose to generate RGB-based synthetic data using a photorealistic visual tool, Unreal Engine, for military target detection in a cross-domain setting. To this end, we conducted synthetic-to-real transfer experiments by training our synthetic dataset and validating on our web-collected real military target datasets. We benchmark the state-of-the-art domain adaptation methods distinguished by the degree of supervision on our proposed train-val dataset pair, and find that current methods using minimal hints on the image (e.g., object class) achieve a substantial improvement over unsupervised or semi-supervised DA methods. From these observations, we recognize the current challenges that remain to be overcome.
Authors: Changgyoon Oh, Jongoh Jeong, Jegyeong Cho, Kuk-Jin Yoon
Abstract: Denoising diffusion probabilistic models have brought tremendous advances in generative tasks, achieving state-of-the-art performance thus far. Current diffusion model-based applications exploit the power of learned visual representations from multistep forward-backward Markovian processes for single-task prediction tasks by attaching a task-specific decoder. However, the heuristic selection of diffusion timestep features still heavily relies on empirical intuition, often leading to sub-optimal performance biased towards certain tasks. To alleviate this constraint, we investigate the significance of versatile diffusion timestep features by adaptively selecting timesteps best suited for the few-shot dense prediction task, evaluated on an arbitrary unseen task. To this end, we propose two modules: Task-aware Timestep Selection (TTS) to select ideal diffusion timesteps based on timestep-wise losses and similarity scores, and Timestep Feature Consolidation (TFC) to consolidate the selected timestep features to improve the dense predictive performance in a few-shot setting. Accompanied by our parameter-efficient fine-tuning adapter, our framework effectively achieves superiority in dense prediction performance given only a few support queries. We empirically validate our learnable timestep consolidation method on the large-scale challenging Taskonomy dataset for dense prediction, particularly for practical universal and few-shot learning scenarios.
Authors: Jongoh Jeong, Taek-Jin Song, Jong-Hwan Kim, Kuk-Jin Yoon
Abstract: Understanding road scenes for visual perception remains crucial for intelligent self-driving cars. In particular, it is desirable to detect unexpected small road hazards reliably in real-time, especially under varying adverse conditions (e.g., weather and daylight). However, existing road driving datasets provide large-scale images acquired in either normal or adverse scenarios only, and often do not contain the road obstacles captured in the same visual domain as for the other classes. To address this, we introduce a new dataset called AVOID, the Adverse Visual Conditions Dataset, for real-time obstacle detection collected in a simulated environment. AVOID consists of a large set of unexpected road obstacles located along each path captured under various weather and time conditions. Each image is coupled with the corresponding semantic and depth maps, raw and semantic LiDAR data, and waypoints, thereby supporting most visual perception tasks. We benchmark the results on high-performing real-time networks for the obstacle detection task, and also propose and conduct ablation studies using a comprehensive multi-task network for semantic segmentation, depth and waypoint prediction tasks.
Authors: Shiqi Dai, Zizhi Ma, Zhicong Luo, Xuesong Yang, Yibin Huang, Wanyue Zhang, Chi Chen, Zonghao Guo, Wang Xu, Yufei Sun, Maosong Sun
Abstract: While Multimodal Large Language Models (MLLMs) have exhibited remarkable general intelligence across diverse domains, their potential in low-altitude applications dominated by Unmanned Aerial Vehicles (UAVs) remains largely underexplored. Existing MLLM benchmarks rarely cover the unique challenges of low-altitude scenarios, while UAV-related evaluations mainly focus on specific tasks such as localization or navigation, without a unified evaluation of MLLMs'general intelligence. To bridge this gap, we present MM-UAVBench, a comprehensive benchmark that systematically evaluates MLLMs across three core capability dimensions-perception, cognition, and planning-in low-altitude UAV scenarios. MM-UAVBench comprises 19 sub-tasks with over 5.7K manually annotated questions, all derived from real-world UAV data collected from public datasets. Extensive experiments on 16 open-source and proprietary MLLMs reveal that current models struggle to adapt to the complex visual and cognitive demands of low-altitude scenarios. Our analyses further uncover critical bottlenecks such as spatial bias and multi-view understanding that hinder the effective deployment of MLLMs in UAV scenarios. We hope MM-UAVBench will foster future research on robust and reliable MLLMs for real-world UAV intelligence.
Authors: Youngchae Kwon, Jinyoung Choi, Injung Kim
Abstract: Fashion item detection is challenging due to the ambiguities introduced by the highly diverse appearances of fashion items and the similarities among item subcategories. To address this challenge, we propose a novel Holistic Detection Transformer (Holi-DETR) that detects fashion items in outfit images holistically, by leveraging contextual information. Fashion items often have meaningful relationships as they are combined to create specific styles. Unlike conventional detectors that detect each item independently, Holi-DETR detects multiple items while reducing ambiguities by leveraging three distinct types of contextual information: (1) the co-occurrence relationship between fashion items, (2) the relative position and size based on inter-item spatial arrangements, and (3) the spatial relationships between items and human body key-points. %Holi-DETR explicitly incorporates three types of contextual information: (1) the co-occurrence probability between fashion items, (2) the relative position and size based on inter-item spatial arrangements, and (3) the spatial relationships between items and human body key-points. To this end, we propose a novel architecture that integrates these three types of heterogeneous contextual information into the Detection Transformer (DETR) and its subsequent models. In experiments, the proposed methods improved the performance of the vanilla DETR and the more recently developed Co-DETR by 3.6 percent points (pp) and 1.1 pp, respectively, in terms of average precision (AP).
Authors: Jiaxu Zhang, Tianshu Hu, Yuan Zhang, Zenan Li, Linjie Luo, Guosheng Lin, Xin Chen
Abstract: Existing AI-driven video creation systems typically treat script drafting and key-shot design as two disjoint tasks: the former relies on large language models, while the latter depends on image generation models. We argue that these two tasks should be unified within a single framework, as logical reasoning and imaginative thinking are both fundamental qualities of a film director. In this work, we propose UniMAGE, a unified director model that bridges user prompts with well-structured scripts, thereby empowering non-experts to produce long-context, multi-shot films by leveraging existing audio-video generation models. To achieve this, we employ the Mixture-of-Transformers architecture that unifies text and image generation. To further enhance narrative logic and keyframe consistency, we introduce a ``first interleaving, then disentangling'' training paradigm. Specifically, we first perform Interleaved Concept Learning, which utilizes interleaved text-image data to foster the model's deeper understanding and imaginative interpretation of scripts. We then conduct Disentangled Expert Learning, which decouples script writing from keyframe generation, enabling greater flexibility and creativity in storytelling. Extensive experiments demonstrate that UniMAGE achieves state-of-the-art performance among open-source models, generating logically coherent video scripts and visually consistent keyframe images.
Authors: Sungho Kang, Hyunkyu Park, Yeonho Lee, Hanbyul Lee, Mijoo Jeong, YeongHyeon Park, Injae Lee, Juneho Yi
Abstract: Anomaly detection plays a vital role in industrial manufacturing. Due to the scarcity of real defect images, unsupervised approaches that rely solely on normal images have been extensively studied. Recently, diffusion-based generative models brought attention to training data synthesis as an alternative solution. In this work, we focus on a strategy to effectively leverage synthetic images to maximize the anomaly detection performance. Previous synthesis strategies are broadly categorized into two groups, presenting a clear trade-off. Rule-based synthesis, such as injecting noise or pasting patches, is cost-effective but often fails to produce realistic defect images. On the other hand, generative model-based synthesis can create high-quality defect images but requires substantial cost. To address this problem, we propose a novel framework that leverages a pre-trained text-guided image-to-image translation model and image retrieval model to efficiently generate synthetic defect images. Specifically, the image retrieval model assesses the similarity of the generated images to real normal images and filters out irrelevant outputs, thereby enhancing the quality and relevance of the generated defect images. To effectively leverage synthetic images, we also introduce a two stage training strategy. In this strategy, the model is first pre-trained on a large volume of images from rule-based synthesis and then fine-tuned on a smaller set of high-quality images. This method significantly reduces the cost for data collection while improving the anomaly detection performance. Experiments on the MVTec AD dataset demonstrate the effectiveness of our approach.
Authors: Minwoo Kim, Hongki Lim
Abstract: Diffusion models have emerged as powerful learned priors for solving inverse problems. However, current iterative solving approaches which alternate between diffusion sampling and data consistency steps typically require hundreds or thousands of steps to achieve high quality reconstruction due to accumulated errors. We address this challenge with SURE Guided Posterior Sampling (SGPS), a method that corrects sampling trajectory deviations using Stein's Unbiased Risk Estimate (SURE) gradient updates and PCA based noise estimation. By mitigating noise induced errors during the critical early and middle sampling stages, SGPS enables more accurate posterior sampling and reduces error accumulation. This allows our method to maintain high reconstruction quality with fewer than 100 Neural Function Evaluations (NFEs). Our extensive evaluation across diverse inverse problems demonstrates that SGPS consistently outperforms existing methods at low NFE counts.
Authors: Dongsheng Li, Chaobo Chen, Siling Wang, Song Gao
Abstract: Detecting infrared gas leaks is critical for environmental monitoring and industrial safety, yet remains difficult because plumes are faint, small, semitransparent, and have weak, diffuse boundaries. We present physics-edge hybrid gas dynamic routing network (PEG-DRNet). First, we introduce the Gas Block, a diffusion-convection unit modeling gas transport: a local branch captures short-range variations, while a large-kernel branch captures long-range propagation. An edge-gated learnable fusion module balances local detail and global context, strengthening weak-contrast plume and contour cues. Second, we propose the adaptive gradient and phase edge operator (AGPEO), computing reliable edge priors from multi-directional gradients and phase-consistent responses. These are transformed by a multi-scale edge perception module (MSEPM) into hierarchical edge features that reinforce boundaries. Finally, the content-adaptive sparse routing path aggregation network (CASR-PAN), with adaptive information modulation modules for fusion and self, selectively propagates informative features across scales based on edge and content cues, improving cross-scale discriminability while reducing redundancy. Experiments on the IIG dataset show that PEG-DRNet achieves an overall AP of 29.8\%, an AP$_{50}$ of 84.3\%, and a small-object AP of 25.3\%, surpassing the RT-DETR-R18 baseline by 3.0\%, 6.5\%, and 5.3\%, respectively, while requiring only 43.7 Gflops and 14.9 M parameters. The proposed PEG-DRNet achieves superior overall performance with the best balance of accuracy and computational efficiency, outperforming existing CNN and Transformer detectors in AP and AP$_{50}$ on the IIG and LangGas dataset.
Authors: Fan Wei, Runmin Dong, Yushan Lai, Yixiang Yang, Zhaoyang Luo, Jinxiao Zhang, Miao Yang, Shuai Yuan, Jiyao Zhao, Bin Luo, Haohuan Fu
Abstract: Diffusion-based remote sensing (RS) generative foundation models are cruial for downstream tasks. However, these models rely on large amounts of globally representative data, which often contain redundancy, noise, and class imbalance, reducing training efficiency and preventing convergence. Existing RS diffusion foundation models typically aggregate multiple classification datasets or apply simplistic deduplication, overlooking the distributional requirements of generation modeling and the heterogeneity of RS imagery. To address these limitations, we propose a training-free, two-stage data pruning approach that quickly select a high-quality subset under high pruning ratios, enabling a preliminary foundation model to converge rapidly and serve as a versatile backbone for generation, downstream fine-tuning, and other applications. Our method jointly considers local information content with global scene-level diversity and representativeness. First, an entropy-based criterion efficiently removes low-information samples. Next, leveraging RS scene classification datasets as reference benchmarks, we perform scene-aware clustering with stratified sampling to improve clustering effectiveness while reducing computational costs on large-scale unlabeled data. Finally, by balancing cluster-level uniformity and sample representativeness, the method enables fine-grained selection under high pruning ratios while preserving overall diversity and representativeness. Experiments show that, even after pruning 85\% of the training data, our method significantly improves convergence and generation quality. Furthermore, diffusion foundation models trained with our method consistently achieve state-of-the-art performance across downstream tasks, including super-resolution and semantic image synthesis. This data pruning paradigm offers practical guidance for developing RS generative foundation models.
Authors: Siyu Zhang, Ying Chen, Lianlei Shan, Runhe Qiu
Abstract: Multimodal fusion of remote sensing images serves as a core technology for overcoming the limitations of single-source data and improving the accuracy of surface information extraction, which exhibits significant application value in fields such as environmental monitoring and urban planning. To address the deficiencies of existing methods, including the failure of fixed resolutions to balance efficiency and detail, as well as the lack of semantic hierarchy in single-scale alignment, this study proposes a Vision-language Model (VLM) framework integrated with two key innovations: the Dynamic Resolution Input Strategy (DRIS) and the Multi-scale Vision-language Alignment Mechanism (MS-VLAM).Specifically, the DRIS adopts a coarse-to-fine approach to adaptively allocate computational resources according to the complexity of image content, thereby preserving key fine-grained features while reducing redundant computational overhead. The MS-VLAM constructs a three-tier alignment mechanism covering object, local-region and global levels, which systematically captures cross-modal semantic consistency and alleviates issues of semantic misalignment and granularity imbalance.Experimental results on the RS-GPT4V dataset demonstrate that the proposed framework significantly improves the accuracy of semantic understanding and computational efficiency in tasks including image captioning and cross-modal retrieval. Compared with conventional methods, it achieves superior performance in evaluation metrics such as BLEU-4 and CIDEr for image captioning, as well as R@10 for cross-modal retrieval. This technical framework provides a novel approach for constructing efficient and robust multimodal remote sensing systems, laying a theoretical foundation and offering technical guidance for the engineering application of intelligent remote sensing interpretation.
Authors: Xingwei Ma, Shiyang Feng, Bo Zhang, Bin Wang
Abstract: Remote sensing change detection (RSCD), a complex multi-image inference task, traditionally uses pixel-based operators or encoder-decoder networks that inadequately capture high-level semantics and are vulnerable to non-semantic perturbations. Although recent multimodal and vision-language model (VLM)-based approaches enhance semantic understanding of change regions by incorporating textual descriptions, they still suffer from challenges such as inaccurate spatial localization, imprecise pixel-level boundary delineation, and limited interpretability. To address these issues, we propose ViLaCD-R1, a two-stage framework comprising a Multi-Image Reasoner (MIR) and a Mask-Guided Decoder (MGD). Specifically, the VLM is trained through supervised fine-tuning (SFT) and reinforcement learning (RL) on block-level dual-temporal inference tasks, taking dual-temporal image patches as input and outputting a coarse change mask. Then, the decoder integrates dual-temporal image features with this coarse mask to predict a precise binary change map. Comprehensive evaluations on multiple RSCD benchmarks demonstrate that ViLaCD-R1 substantially improves true semantic change recognition and localization, robustly suppresses non-semantic variations, and achieves state-of-the-art accuracy in complex real-world scenarios.
Authors: Shin seong Kim, Minjung Shin, Hyunin Cho, Youngjung Uh
Abstract: Recent text-to-image diffusion models have significantly improved visual quality and text alignment. However, generating a sequence of images while preserving consistent character identity across diverse scene descriptions remains a challenging task. Existing methods often struggle with a trade-off between maintaining identity consistency and ensuring per-image prompt alignment. In this paper, we introduce a novel framework, ASemconsist, that addresses this challenge through selective text embedding modification, enabling explicit semantic control over character identity without sacrificing prompt alignment. Furthermore, based on our analysis of padding embeddings in FLUX, we propose a semantic control strategy that repurposes padding embeddings as semantic containers. Additionally, we introduce an adaptive feature-sharing strategy that automatically evaluates textual ambiguity and applies constraints only to the ambiguous identity prompt. Finally, we propose a unified evaluation protocol, the Consistency Quality Score (CQS), which integrates identity preservation and per-image text alignment into a single comprehensive metric, explicitly capturing performance imbalances between the two metrics. Our framework achieves state-of-the-art performance, effectively overcoming prior trade-offs. Project page: https://minjung-s.github.io/asemconsist
Authors: Masaya Takabe, Hiroshi Watanabe, Sujun Hong, Tomohiro Ikai, Zheming Fan, Ryo Ishimoto, Kakeru Sugimoto, Ruri Imichi
Abstract: Image representation is a fundamental task in computer vision. Recently, Gaussian Splatting has emerged as an efficient representation framework, and its extension to 2D image representation enables lightweight, yet expressive modeling of visual content. While recent 2D Gaussian Splatting (2DGS) approaches provide compact storage and real-time decoding, they often produce blurry or indistinct boundaries when the number of Gaussians is small due to the lack of contour awareness. In this work, we propose a Contour Information-Aware 2D Gaussian Splatting framework that incorporates object segmentation priors into Gaussian-based image representation. By constraining each Gaussian to a specific segmentation region during rasterization, our method prevents cross-boundary blending and preserves edge structures under high compression. We also introduce a warm-up scheme to stabilize training and improve convergence. Experiments on synthetic color charts and the DAVIS dataset demonstrate that our approach achieves higher reconstruction quality around object edges compared to existing 2DGS methods. The improvement is particularly evident in scenarios with very few Gaussians, while our method still maintains fast rendering and low memory usage.
Authors: Tong Shao, Yusen Fu, Guoying Sun, Jingde Kong, Zhuotao Tian, Jingyong Su
Abstract: Although Diffusion Transformer (DiT) has emerged as a predominant architecture for image and video generation, its iterative denoising process results in slow inference, which hinders broader applicability and development. Caching-based methods achieve training-free acceleration, while suffering from considerable computational error. Existing methods typically incorporate error correction strategies such as pruning or prediction to mitigate it. However, their fixed caching strategy fails to adapt to the complex error variations during denoising, which limits the full potential of error correction. To tackle this challenge, we propose a novel fidelity-optimization plugin for existing error correction methods via cumulative error minimization, named CEM. CEM predefines the error to characterize the sensitivity of model to acceleration jointly influenced by timesteps and cache intervals. Guided by this prior, we formulate a dynamic programming algorithm with cumulative error approximation for strategy optimization, which achieves the caching error minimization, resulting in a substantial improvement in generation fidelity. CEM is model-agnostic and exhibits strong generalization, which is adaptable to arbitrary acceleration budgets. It can be seamlessly integrated into existing error correction frameworks and quantized models without introducing any additional computational overhead. Extensive experiments conducted on nine generation models and quantized methods across three tasks demonstrate that CEM significantly improves generation fidelity of existing acceleration models, and outperforms the original generation performance on FLUX.1-dev, PixArt-$\alpha$, StableDiffusion1.5 and Hunyuan. The code will be made publicly available.
Authors: Xu Lin, Jinlong Peng, Zhenye Gan, Jiawen Zhu, Jun Liu
Abstract: Existing Real-Time Object Detection (RTOD) methods commonly adopt YOLO-like architectures for their favorable trade-off between accuracy and speed. However, these models rely on static dense computation that applies uniform processing to all inputs, misallocating representational capacity and computational resources such as over-allocating on trivial scenes while under-serving complex ones. This mismatch results in both computational redundancy and suboptimal detection performance. To overcome this limitation, we propose YOLO-Master, a novel YOLO-like framework that introduces instance-conditional adaptive computation for RTOD. This is achieved through a Efficient Sparse Mixture-of-Experts (ES-MoE) block that dynamically allocates computational resources to each input according to its scene complexity. At its core, a lightweight dynamic routing network guides expert specialization during training through a diversity enhancing objective, encouraging complementary expertise among experts. Additionally, the routing network adaptively learns to activate only the most relevant experts, thereby improving detection performance while minimizing computational overhead during inference. Comprehensive experiments on five large-scale benchmarks demonstrate the superiority of YOLO-Master. On MS COCO, our model achieves 42.4% AP with 1.62ms latency, outperforming YOLOv13-N by +0.8% mAP and 17.8% faster inference. Notably, the gains are most pronounced on challenging dense scenes, while the model preserves efficiency on typical inputs and maintains real-time inference speed. Code will be available.
Authors: Arman Martirosyan, Shahane Tigranyan, Maria Razzhivina, Artak Aslanyan, Nazgul Salikhova, Ilya Makarov, Andrey Savchenko, Aram Avetisyan
Abstract: Micro-gesture recognition and behavior-based emotion prediction are both highly challenging tasks that require modeling subtle, fine-grained human behaviors, primarily leveraging video and skeletal pose data. In this work, we present two multimodal frameworks designed to tackle both problems on the iMiGUE dataset. For micro-gesture classification, we explore the complementary strengths of RGB and 3D pose-based representations to capture nuanced spatio-temporal patterns. To comprehensively represent gestures, video, and skeletal embeddings are extracted using MViTv2-S and 2s-AGCN, respectively. Then, they are integrated through a Cross-Modal Token Fusion module to combine spatial and pose information. For emotion recognition, our framework extends to behavior-based emotion prediction, a binary classification task identifying emotional states based on visual cues. We leverage facial and contextual embeddings extracted using SwinFace and MViTv2-S models and fuse them through an InterFusion module designed to capture emotional expressions and body gestures. Experiments conducted on the iMiGUE dataset, within the scope of the MiGA 2025 Challenge, demonstrate the robust performance and accuracy of our method in the behavior-based emotion prediction task, where our approach secured 2nd place.
Authors: Md. Sazzadul Islam Prottasha, Nabil Walid Rafi
Abstract: Multimodal Large Language Models (LLMs) introduce an emerging paradigm for medical imaging by interpreting scans through the lens of extensive clinical knowledge, offering a transformative approach to disease classification. This study presents a critical comparison between two fundamentally different AI architectures: the specialized open-source agent MedGemma and the proprietary large multimodal model GPT-4 for diagnosing six different diseases. The MedGemma-4b-it model, fine-tuned using Low-Rank Adaptation (LoRA), demonstrated superior diagnostic capability by achieving a mean test accuracy of 80.37% compared to 69.58% for the untuned GPT-4. Furthermore, MedGemma exhibited notably higher sensitivity in high-stakes clinical tasks, such as cancer and pneumonia detection. Quantitative analysis via confusion matrices and classification reports provides comprehensive insights into model performance across all categories. These results emphasize that domain-specific fine-tuning is essential for minimizing hallucinations in clinical implementation, positioning MedGemma as a sophisticated tool for complex, evidence-based medical reasoning.
Authors: Ke Niu, Haiyang Yu, Zhuofan Chen, Zhengtao Yao, Weitao Jia, Xiaodong Ge, Jingqun Tang, Benlei Cui, Bin Li, Xiangyang Xue
Abstract: Computer-Aided Design (CAD) is essential in industrial design, but the complexity of traditional CAD modeling and workflows presents significant challenges for automating the generation of high-precision, editable CAD models. Existing methods that reconstruct 3D models from sketches often produce non-editable and approximate models that fall short of meeting the stringent requirements for precision and editability in industrial design. Moreover, the reliance on text or image-based inputs often requires significant manual annotation, limiting their scalability and applicability in industrial settings. To overcome these challenges, we propose the Heterogeneous Collaborative Multi-Expert Reinforcement Learning (CME-CAD) paradigm, a novel training paradigm for CAD code generation. Our approach integrates the complementary strengths of these models, facilitating collaborative learning and improving the model's ability to generate accurate, constraint-compatible, and fully editable CAD models. We introduce a two-stage training process: Multi-Expert Fine-Tuning (MEFT), and Multi-Expert Reinforcement Learning (MERL). Additionally, we present CADExpert, an open-source benchmark consisting of 17,299 instances, including orthographic projections with precise dimension annotations, expert-generated Chain-of-Thought (CoT) processes, executable CADQuery code, and rendered 3D models.
Authors: Xiu Li
Abstract: We study visual representation learning from a structural and topological perspective. We begin from a single hypothesis: that visual understanding presupposes a semantic language for vision, in which many perceptual observations correspond to a small number of discrete semantic states. Together with widely assumed premises on transferability and abstraction in representation learning, this hypothesis implies that the visual observation space must be organized in a fiber bundle like structure, where nuisance variation populates fibers and semantics correspond to a quotient base space. From this structure we derive two theoretical consequences. First, the semantic quotient $X/G$ is not a submanifold of $X$ and cannot be obtained through smooth deformation alone, semantic invariance requires a non-homeomorphic, discriminative target, for example, supervision via labels, cross instance identification, or multimodal alignment that supplies explicit semantic equivalence. Second, we show that approximating the quotient also places structural demands on the model architecture. Semantic abstraction requires not only an external semantic target, but a representation mechanism capable of supporting topology change: an expand-and-snap process in which the manifold is first geometrically expanded to separate structure and then collapsed to form discrete semantic regions. We emphasize that these results are interpretive rather than prescriptive: the framework provides a topological lens that aligns with empirical regularities observed in large-scale discriminative and multimodal models, and with classical principles in statistical learning theory.
Authors: Niki Amini-Naieni, Andrew Zisserman
Abstract: The flexibility and accuracy of methods for automatically counting objects in images and videos are limited by the way the object can be specified. While existing methods allow users to describe the target object with text and visual examples, the visual examples must be manually annotated inside the image, and there is no way to specify what not to count. To address these gaps, we introduce novel capabilities that expand how the target object can be specified. Specifically, we extend the prompt to enable what not to count to be described with text and/or visual examples, introduce the concept of `pseudo-exemplars' that automate the annotation of visual examples at inference, and extend counting models to accept visual examples from both natural and synthetic external images. We also use our new counting model, CountGD++, as a vision expert agent for an LLM. Together, these contributions expand the prompt flexibility of multi-modal open-world counting and lead to significant improvements in accuracy, efficiency, and generalization across multiple datasets. Code is available at https://github.com/niki-amini-naieni/CountGDPlusPlus.
Authors: Kanghee Lee, Injae Lee, Minseok Kwak, Kwonyoung Ryu, Jungi Hong, Jaesik Park
Abstract: The rapid progress of Multimodal Large Language Models (MLLMs) has unlocked the potential for enhanced 3D scene understanding and spatial reasoning. However, existing approaches often rely on pre-constructed 3D representations or off-the-shelf reconstruction pipelines, which constrain scalability and real-world applicability. A recent line of work explores learning spatial reasoning directly from multi-view images, enabling Vision-Language Models (VLMs) to understand 3D scenes without explicit 3D reconstructions. Nevertheless, key challenges that frequently arise in real-world environments, such as partial visibility, occlusion, and low-overlap conditions that require spatial reasoning from fragmented visual cues, remain under-explored. To address these limitations, we propose a scalable multi-view data generation and annotation pipeline that constructs realistic spatial reasoning QAs, resulting in SpatialMosaic, a comprehensive instruction-tuning dataset featuring 2M QA pairs. We further introduce SpatialMosaic-Bench, a challenging benchmark for evaluating multi-view spatial reasoning under realistic and challenging scenarios, consisting of 1M QA pairs across 6 tasks. In addition, we present SpatialMosaicVLM, a hybrid framework that integrates 3D reconstruction models as geometry encoders within VLMs for robust spatial reasoning. Extensive experiments demonstrate that our proposed dataset and VQA tasks effectively enhance spatial reasoning under challenging multi-view conditions, validating the effectiveness of our data generation pipeline in constructing realistic and diverse QA pairs. Code and dataset will be available soon.
Authors: Shuyuan Lin, Mengtin Lo, Haosheng Chen, Yanjie Liang, Qiangqiang Wu
Abstract: Two-view correspondence learning is a key task in computer vision, which aims to establish reliable matching relationships for applications such as camera pose estimation and 3D reconstruction. However, existing methods have limitations in local geometric modeling and cross-stage information optimization, which make it difficult to accurately capture the geometric constraints of matched pairs and thus reduce the robustness of the model. To address these challenges, we propose a Multi-Graph Contextual Attention Network (MGCA-Net), which consists of a Contextual Geometric Attention (CGA) module and a Cross-Stage Multi-Graph Consensus (CSMGC) module. Specifically, CGA dynamically integrates spatial position and feature information via an adaptive attention mechanism and enhances the capability to capture both local and global geometric relationships. Meanwhile, CSMGC establishes geometric consensus via a cross-stage sparse graph network, ensuring the consistency of geometric information across different stages. Experimental results on two representative YFCC100M and SUN3D datasets show that MGCA-Net significantly outperforms existing SOTA methods in the outlier rejection and camera pose estimation tasks. Source code is available at http://www.linshuyuan.com.
Authors: Yifei Li, Haoyuan He, Yu Zheng, Bingyao Yu, Wenzhao Zheng, Lei Chen, Jie Zhou, Jiwen Lu
Abstract: The accessibility surge and abuse risks of user-friendly image editing models have created an urgent need for generalizable, up-to-date methods for Image Manipulation Detection and Localization (IMDL). Current IMDL research typically uses cross-dataset evaluation, where models trained on one benchmark are tested on others. However, this simplified evaluation approach conceals the fragility of existing methods when handling diverse AI-generated content, leading to misleading impressions of progress. This paper challenges this illusion by proposing NeXT-IMDL, a large-scale diagnostic benchmark designed not just to collect data, but to probe the generalization boundaries of current detectors systematically. Specifically, NeXT-IMDL categorizes AIGC-based manipulations along four fundamental axes: editing models, manipulation types, content semantics, and forgery granularity. Built upon this, NeXT-IMDL implements five rigorous cross-dimension evaluation protocols. Our extensive experiments on 11 representative models reveal a critical insight: while these models perform well in their original settings, they exhibit systemic failures and significant performance degradation when evaluated under our designed protocols that simulate real-world, various generalization scenarios. By providing this diagnostic toolkit and the new findings, we aim to advance the development towards building truly robust, next-generation IMDL models.
Authors: Le Shen, Qiao Qian, Tan Yu, Ke Zhou, Tianhang Yu, Yu Zhan, Zhenjie Wang, Ming Tao, Shunshun Yin, Siyuan Liu
Abstract: Deploying massive diffusion models for real-time, infinite-duration, audio-driven avatar generation presents a significant engineering challenge, primarily due to the conflict between computational load and strict latency constraints. Existing approaches often compromise visual fidelity by enforcing strictly unidirectional attention mechanisms or reducing model capacity. To address this problem, we introduce \textbf{SoulX-LiveTalk}, a 14B-parameter framework optimized for high-fidelity real-time streaming. Diverging from conventional unidirectional paradigms, we use a \textbf{Self-correcting Bidirectional Distillation} strategy that retains bidirectional attention within video chunks. This design preserves critical spatiotemporal correlations, significantly enhancing motion coherence and visual detail. To ensure stability during infinite generation, we incorporate a \textbf{Multi-step Retrospective Self-Correction Mechanism}, enabling the model to autonomously recover from accumulated errors and preventing collapse. Furthermore, we engineered a full-stack inference acceleration suite incorporating hybrid sequence parallelism, Parallel VAE, and kernel-level optimizations. Extensive evaluations confirm that SoulX-LiveTalk is the first 14B-scale system to achieve a \textbf{sub-second start-up latency (0.87s)} while reaching a real-time throughput of \textbf{32 FPS}, setting a new standard for high-fidelity interactive digital human synthesis.
Authors: Xiaolan Li, Wanquan Liu, Pengcheng Li, Pengyu Jie, Chenqiang Gao
Abstract: Three-dimensional (3D) tooth instance segmentation remains challenging due to crowded arches, ambiguous tooth-gingiva boundaries, missing teeth, and rare yet clinically important third molars. Native 3D methods relying on geometric cues often suffer from boundary leakage, center drift, and inconsistent tooth identities, especially for minority classes and complex anatomies. Meanwhile, 2D foundation models such as the Segment Anything Model (SAM) provide strong boundary-aware semantics, but directly applying them in 3D is impractical in clinical workflows. To address these issues, we propose SOFTooth, a semantics-enhanced, order-aware 2D-3D fusion framework that leverages frozen 2D semantics without explicit 2D mask supervision. First, a point-wise residual gating module injects occlusal-view SAM embeddings into 3D point features to refine tooth-gingiva and inter-tooth boundaries. Second, a center-guided mask refinement regularizes consistency between instance masks and geometric centroids, reducing center drift. Furthermore, an order-aware Hungarian matching strategy integrates anatomical tooth order and center distance into similarity-based assignment, ensuring coherent labeling even under missing or crowded dentitions. On 3DTeethSeg'22, SOFTooth achieves state-of-the-art overall accuracy and mean IoU, with clear gains on cases involving third molars, demonstrating that rich 2D semantics can be effectively transferred to 3D tooth instance segmentation without 2D fine-tuning.
Authors: Henglin Liu, Nisha Huang, Chang Liu, Jiangpeng Yan, Huijuan Huang, Jixuan Ying, Tong-Yee Lee, Pengfei Wan, Xiangyang Ji
Abstract: The aesthetic quality assessment task is crucial for developing a human-aligned quantitative evaluation system for AIGC. However, its inherently complex nature, spanning visual perception, cognition, and emotion, poses fundamental challenges. Although aesthetic descriptions offer a viable representation of this complexity, two critical challenges persist: (1) data scarcity and imbalance: existing dataset overly focuses on visual perception and neglects deeper dimensions due to the expensive manual annotation; and (2) model fragmentation: current visual networks isolate aesthetic attributes with multi-branch encoder, while multimodal methods represented by contrastive learning struggle to effectively process long-form textual descriptions. To resolve challenge (1), we first present the Refined Aesthetic Description (RAD) dataset, a large-scale (70k), multi-dimensional structured dataset, generated via an iterative pipeline without heavy annotation costs and easy to scale. To address challenge (2), we propose ArtQuant, an aesthetics assessment framework for artistic images which not only couples isolated aesthetic dimensions through joint description generation, but also better models long-text semantics with the help of LLM decoders. Besides, theoretical analysis confirms this symbiosis: RAD's semantic adequacy (data) and generation paradigm (model) collectively minimize prediction entropy, providing mathematical grounding for the framework. Our approach achieves state-of-the-art performance on several datasets while requiring only 33% of conventional training epochs, narrowing the cognitive gap between artistic images and aesthetic judgment. We will release both code and dataset to support future research.
Authors: Tianze Xia, Yongkang Li, Lijun Zhou, Jingfeng Yao, Kaixin Xiong, Haiyang Sun, Bing Wang, Kun Ma, Hangjun Ye, Wenyu Liu, Xinggang Wang
Abstract: World models have become crucial for autonomous driving, as they learn how scenarios evolve over time to address the long-tail challenges of the real world. However, current approaches relegate world models to limited roles: they operate within ostensibly unified architectures that still keep world prediction and motion planning as decoupled processes. To bridge this gap, we propose DriveLaW, a novel paradigm that unifies video generation and motion planning. By directly injecting the latent representation from its video generator into the planner, DriveLaW ensures inherent consistency between high-fidelity future generation and reliable trajectory planning. Specifically, DriveLaW consists of two core components: DriveLaW-Video, our powerful world model that generates high-fidelity forecasting with expressive latent representations, and DriveLaW-Act, a diffusion planner that generates consistent and reliable trajectories from the latent of DriveLaW-Video, with both components optimized by a three-stage progressive training strategy. The power of our unified paradigm is demonstrated by new state-of-the-art results across both tasks. DriveLaW not only advances video prediction significantly, surpassing best-performing work by 33.3% in FID and 1.8% in FVD, but also achieves a new record on the NAVSIM planning benchmark.
Authors: Dohyun Kim, Seungwoo Lyu, Seung Wook Kim, Paul Hongsuck Seo
Abstract: Diffusion models have achieved impressive results in generative tasks such as text-to-image synthesis, yet they often struggle to fully align outputs with nuanced user intent and maintain consistent aesthetic quality. Existing preference-based training methods like Diffusion Direct Preference Optimization help address these issues but rely on costly and potentially noisy human-labeled datasets. In this work, we introduce Direct Diffusion Score Preference Optimization (DDSPO), which directly derives per-timestep supervision from winning and losing policies when such policies are available. Unlike prior methods that operate solely on final samples, DDSPO provides dense, transition-level signals across the denoising trajectory. In practice, we avoid reliance on labeled data by automatically generating preference signals using a pretrained reference model: we contrast its outputs when conditioned on original prompts versus semantically degraded variants. This practical strategy enables effective score-space preference supervision without explicit reward modeling or manual annotations. Empirical results demonstrate that DDSPO improves text-image alignment and visual quality, outperforming or matching existing preference-based methods while requiring significantly less supervision. Our implementation is available at: https://dohyun-as.github.io/DDSPO
Authors: Jesse Brouwers, Xiaoyan Xing, Alexander Timans
Abstract: Foundation models for segmentation such as the Segment Anything Model (SAM) family exhibit strong zero-shot performance, but remain vulnerable in shifted or limited-knowledge domains. This work investigates whether uncertainty quantification can mitigate such challenges and enhance model generalisability in a domain-agnostic manner. To this end, we (1) curate UncertSAM, a benchmark comprising eight datasets designed to stress-test SAM under challenging segmentation conditions including shadows, transparency, and camouflage; (2) evaluate a suite of lightweight, post-hoc uncertainty estimation methods; and (3) assess a preliminary uncertainty-guided prediction refinement step. Among evaluated approaches, a last-layer Laplace approximation yields uncertainty estimates that correlate well with segmentation errors, indicating a meaningful signal. While refinement benefits are preliminary, our findings underscore the potential of incorporating uncertainty into segmentation models to support robust, domain-agnostic performance. Our benchmark and code are made publicly available.
Authors: Mustafa Demetgul, Sanja Lazarova Molnar
Abstract: Monitoring states of road surfaces provides valuable information for the planning and controlling vehicles and active vehicle control systems. Classical road monitoring methods are expensive and unsystematic because they require time for measurements. This article proposes an real time system based on weather conditional data and road surface condition data. For this purpose, we collected data with a mobile phone camera on the roads around the campus of the Karlsruhe Institute of Technology. We tested a large number of different image-based deep learning algorithms for road classification. In addition, we used road acceleration data along with road image data for training by using them as images. We compared the performances of acceleration-based and camera image-based approaches. The performances of the simple Alexnet, LeNet, VGG, and Resnet algorithms were compared as deep learning algorithms. For road condition classification, 5 classes were considered: asphalt, damaged asphalt, gravel road, damaged gravel road, pavement road and over 95% accuracy performance was achieved. It is also proposed to use the acceleration or the camera image to classify the road surface according to the weather and the time of day using fuzzy logic.
Authors: Shuhong Liu, Chenyu Bao, Ziteng Cui, Yun Liu, Xuangeng Chu, Lin Gu, Marcos V. Conde, Ryo Umagami, Tomohiro Hashimoto, Zijian Hu, Tianhan Xu, Yuan Gan, Yusuke Kurose, Tatsuya Harada
Abstract: We introduce RealX3D, a real-capture benchmark for multi-view visual restoration and 3D reconstruction under diverse physical degradations. RealX3D groups corruptions into four families, including illumination, scattering, occlusion, and blurring, and captures each at multiple severity levels using a unified acquisition protocol that yields pixel-aligned LQ/GT views. Each scene includes high-resolution capture, RAW images, and dense laser scans, from which we derive world-scale meshes and metric depth. Benchmarking a broad range of optimization-based and feed-forward methods shows substantial degradation in reconstruction quality under physical corruptions, underscoring the fragility of current multi-view pipelines in real-world challenging environments.
Authors: Zongsheng Cao, Yangfan He, Anran Liu, Jun Xie, Feng Chen, Zepeng Wang
Abstract: Large Vision-Language Models (LVLMs) have achieved impressive progress in multi-modal understanding and generation. However, they still tend to produce hallucinated content that is inconsistent with the visual input, which limits their reliability in real-world applications. We propose \textbf{CoFi-Dec}, a training-free decoding framework that mitigates hallucinations by integrating generative self-feedback with coarse-to-fine visual conditioning. Inspired by the human visual process from global scene perception to detailed inspection, CoFi-Dec first generates two intermediate textual responses conditioned on coarse- and fine-grained views of the original image. These responses are then transformed into synthetic images using a text-to-image model, forming multi-level visual hypotheses that enrich grounding cues. To unify the predictions from these multiple visual conditions, we introduce a Wasserstein-based fusion mechanism that aligns their predictive distributions into a geometrically consistent decoding trajectory. This principled fusion reconciles high-level semantic consistency with fine-grained visual grounding, leading to more robust and faithful outputs. Extensive experiments on six hallucination-focused benchmarks show that CoFi-Dec substantially reduces both entity-level and semantic-level hallucinations, outperforming existing decoding strategies. The framework is model-agnostic, requires no additional training, and can be seamlessly applied to a wide range of LVLMs. The implementation is available at https://github.com/AI-Researcher-Team/CoFi-Dec.
Authors: Kayathri Vigneswaran, Hugo Retief, Jai Clifford Holmes, Mariangel Garcia Andarcia, Hansaka Tennakoon
Abstract: Accurate and continuous monitoring of river water levels is essential for flood forecasting, water resource management, and ecological protection. Traditional hydrological observation methods are often limited by manual measurement errors and environmental constraints. This study presents a hybrid framework integrating vision based waterline detection, YOLOv8 pose scale extraction, and large multimodal language models (GPT 4o and Gemini 2.0 Flash) for automated river gauge plate reading. The methodology involves sequential stages of image preprocessing, annotation, waterline detection, scale gap estimation, and numeric reading extraction. Experiments demonstrate that waterline detection achieved high precision of 94.24 percent and an F1 score of 83.64 percent, while scale gap detection provided accurate geometric calibration for subsequent reading extraction. Incorporating scale gap metadata substantially improved the predictive performance of LLMs, with Gemini Stage 2 achieving the highest accuracy, with a mean absolute error of 5.43 cm, root mean square error of 8.58 cm, and R squared of 0.84 under optimal image conditions. Results highlight the sensitivity of LLMs to image quality, with degraded images producing higher errors, and underscore the importance of combining geometric metadata with multimodal artificial intelligence for robust water level estimation. Overall, the proposed approach offers a scalable, efficient, and reliable solution for automated hydrological monitoring, demonstrating potential for real time river gauge digitization and improved water resource management.
Authors: Bohan Xiao, Peiyong Wang, Qisheng He, Ming Dong
Abstract: Image-to-Image (I2I) translation involves converting an image from one domain to another. Deterministic I2I translation, such as in image super-resolution, extends this concept by guaranteeing that each input generates a consistent and predictable output, closely matching the ground truth (GT) with high fidelity. In this paper, we propose a denoising Brownian bridge model with dual approximators (Dual-approx Bridge), a novel generative model that exploits the Brownian bridge dynamics and two neural network-based approximators (one for forward and one for reverse process) to produce faithful output with negligible variance and high image quality in I2I translations. Our extensive experiments on benchmark datasets including image generation and super-resolution demonstrate the consistent and superior performance of Dual-approx Bridge in terms of image quality and faithfulness to GT when compared to both stochastic and deterministic baselines. Project page and code: https://github.com/bohan95/dual-app-bridge
Authors: Yuxin Wen, Qing Shuai, Di Kang, Jing Li, Cheng Wen, Yue Qian, Ningxin Jiao, Changhai Chen, Weijie Chen, Yiran Wang, Jinkun Guo, Dongyue An, Han Liu, Yanyu Tong, Chao Zhang, Qing Guo, Juan Chen, Qiao Zhang, Youyi Zhang, Zihao Yao, Cheng Zhang, Hong Duan, Xiaoping Wu, Qi Chen, Fei Cheng, Liang Dong, Peng He, Hao Zhang, Jiaxin Lin, Chao Zhang, Zhongyi Fan, Yifan Li, Zhichao Hu, Yuhong Liu, Linus, Jie Jiang, Xiaolong Li, Linchao Bao
Abstract: We present HY-Motion 1.0, a series of state-of-the-art, large-scale, motion generation models capable of generating 3D human motions from textual descriptions. HY-Motion 1.0 represents the first successful attempt to scale up Diffusion Transformer (DiT)-based flow matching models to the billion-parameter scale within the motion generation domain, delivering instruction-following capabilities that significantly outperform current open-source benchmarks. Uniquely, we introduce a comprehensive, full-stage training paradigm -- including large-scale pretraining on over 3,000 hours of motion data, high-quality fine-tuning on 400 hours of curated data, and reinforcement learning from both human feedback and reward models -- to ensure precise alignment with the text instruction and high motion quality. This framework is supported by our meticulous data processing pipeline, which performs rigorous motion cleaning and captioning. Consequently, our model achieves the most extensive coverage, spanning over 200 motion categories across 6 major classes. We release HY-Motion 1.0 to the open-source community to foster future research and accelerate the transition of 3D human motion generation models towards commercial maturity.
Authors: Shuyuan Lin, Wenwu Peng, Junjie Huang, Qiang Qi, Miaohui Wang, Jian Weng
Abstract: Robust and discriminative feature learning is critical for high-quality point cloud registration. However, existing deep learning-based methods typically rely on Euclidean neighborhood-based strategies for feature extraction, which struggle to effectively capture the implicit semantics and structural consistency in point clouds. To address these issues, we propose a multi-domain context integration network (MCI-Net) that improves feature representation and registration performance by aggregating contextual cues from diverse domains. Specifically, we propose a graph neighborhood aggregation module, which constructs a global graph to capture the overall structural relationships within point clouds. We then propose a progressive context interaction module to enhance feature discriminability by performing intra-domain feature decoupling and inter-domain context interaction. Finally, we design a dynamic inlier selection method that optimizes inlier weights using residual information from multiple iterations of pose estimation, thereby improving the accuracy and robustness of registration. Extensive experiments on indoor RGB-D and outdoor LiDAR datasets show that the proposed MCI-Net significantly outperforms existing state-of-the-art methods, achieving the highest registration recall of 96.4\% on 3DMatch. Source code is available at http://www.linshuyuan.com.
Authors: Shuyuan Lin, Hailiang Liao, Qiang Qi, Junjie Huang, Taotao Lai, Jian Weng
Abstract: Recent research has focused on using convolutional neural networks (CNNs) as the backbones in two-view correspondence learning, demonstrating significant superiority over methods based on multilayer perceptrons. However, CNN backbones that are not tailored to specific tasks may fail to effectively aggregate global context and oversmooth dense motion fields in scenes with large disparity. To address these problems, we propose a novel network named SC-Net, which effectively integrates bilateral context from both spatial and channel perspectives. Specifically, we design an adaptive focused regularization module (AFR) to enhance the model's position-awareness and robustness against spurious motion samples, thereby facilitating the generation of a more accurate motion field. We then propose a bilateral field adjustment module (BFA) to refine the motion field by simultaneously modeling long-range relationships and facilitating interaction across spatial and channel dimensions. Finally, we recover the motion vectors from the refined field using a position-aware recovery module (PAR) that ensures consistency and precision. Extensive experiments demonstrate that SC-Net outperforms state-of-the-art methods in relative pose estimation and outlier removal tasks on YFCC100M and SUN3D datasets. Source code is available at http://www.linshuyuan.com.
Authors: Zongsheng Cao, Yangfan He, Anran Liu, Feng Chen, Zepeng Wang, Jun Xie
Abstract: Large Video Language Models (LVLMs) have rapidly emerged as the focus of multimedia AI research. Nonetheless, when confronted with lengthy videos, these models struggle: their temporal windows are narrow, and they fail to notice fine-grained semantic shifts that unfold over extended durations. Moreover, mainstream text-based retrieval pipelines, which rely chiefly on surface-level lexical overlap, ignore the rich temporal interdependence among visual, audio, and subtitle channels. To mitigate these limitations, we propose TV-RAG, a training-free architecture that couples temporal alignment with entropy-guided semantics to improve long-video reasoning. The framework contributes two main mechanisms: \emph{(i)} a time-decay retrieval module that injects explicit temporal offsets into the similarity computation, thereby ranking text queries according to their true multimedia context; and \emph{(ii)} an entropy-weighted key-frame sampler that selects evenly spaced, information-dense frames, reducing redundancy while preserving representativeness. By weaving these temporal and semantic signals together, TV-RAG realises a dual-level reasoning routine that can be grafted onto any LVLM without re-training or fine-tuning. The resulting system offers a lightweight, budget-friendly upgrade path and consistently surpasses most leading baselines across established long-video benchmarks such as Video-MME, MLVU, and LongVideoBench, confirming the effectiveness of our model. The code can be found at https://github.com/AI-Researcher-Team/TV-RAG.
Authors: Mingyuan Jiu, Hailong Zhu, Wenchuan Wei, Hichem Sahbi, Rongrong Ji, Mingliang Xu
Abstract: Context modeling is crucial for visual recognition, enabling highly discriminative image representations by integrating both intrinsic and extrinsic relationships between objects and labels in images. A limitation in current approaches is their focus on basic geometric relationships or localized features, often neglecting cross-scale contextual interactions between objects. This paper introduces the Deep Panoptic Context Aggregation Network (PanCAN), a novel approach that hierarchically integrates multi-order geometric contexts through cross-scale feature aggregation in a high-dimensional Hilbert space. Specifically, PanCAN learns multi-order neighborhood relationships at each scale by combining random walks with an attention mechanism. Modules from different scales are cascaded, where salient anchors at a finer scale are selected and their neighborhood features are dynamically fused via attention. This enables effective cross-scale modeling that significantly enhances complex scene understanding by combining multi-order and cross-scale context-aware features. Extensive multi-label classification experiments on NUS-WIDE, PASCAL VOC2007, and MS-COCO benchmarks demonstrate that PanCAN consistently achieves competitive results, outperforming state-of-the-art techniques in both quantitative and qualitative evaluations, thereby substantially improving multi-label classification performance.
Authors: Donghao Zhou, Jingyu Lin, Guibao Shen, Quande Liu, Jialin Gao, Lihao Liu, Lan Du, Cunjian Chen, Chi-Wing Fu, Xiaowei Hu, Pheng-Ann Heng
Abstract: Recent visual generative models enable story generation with consistent characters from text, but human-centric story generation faces additional challenges, such as maintaining detailed and diverse human face consistency and coordinating multiple characters across different images. This paper presents IdentityStory, a framework for human-centric story generation that ensures consistent character identity across multiple sequential images. By taming identity-preserving generators, the framework features two key components: Iterative Identity Discovery, which extracts cohesive character identities, and Re-denoising Identity Injection, which re-denoises images to inject identities while preserving desired context. Experiments on the ConsiStory-Human benchmark demonstrate that IdentityStory outperforms existing methods, particularly in face consistency, and supports multi-character combinations. The framework also shows strong potential for applications such as infinite-length story generation and dynamic character composition.
Authors: Hexin Zhang, Dong Li, Jie Huang, Bingzhou Wang, Xueyang Fu, Zhengjun Zha
Abstract: Diffusion models have become a leading paradigm for image super-resolution (SR), but existing methods struggle to guarantee both the high-frequency perceptual quality and the low-frequency structural fidelity of generated images. Although inference-time scaling can theoretically improve this trade-off by allocating more computation, existing strategies remain suboptimal: reward-driven particle optimization often causes perceptual over-smoothing, while optimal-path search tends to lose structural consistency. To overcome these difficulties, we propose Iterative Diffusion Inference-Time Scaling with Adaptive Frequency Steering (IAFS), a training-free framework that jointly leverages iterative refinement and frequency-aware particle fusion. IAFS addresses the challenge of balancing perceptual quality and structural fidelity by progressively refining the generated image through iterative correction of structural deviations. Simultaneously, it ensures effective frequency fusion by adaptively integrating high-frequency perceptual cues with low-frequency structural information, allowing for a more accurate and balanced reconstruction across different image details. Extensive experiments across multiple diffusion-based SR models show that IAFS effectively resolves the perception-fidelity conflict, yielding consistently improved perceptual detail and structural accuracy, and outperforming existing inference-time scaling methods.
Authors: Binhe Yu, Zhen Wang, Kexin Li, Yuqian Yuan, Wenqiao Zhang, Long Chen, Juncheng Li, Jun Xiao, Yueting Zhuang
Abstract: Multi-subject customization aims to synthesize multiple user-specified subjects into a coherent image. To address issues such as subjects missing or conflicts, recent works incorporate layout guidance to provide explicit spatial constraints. However, existing methods still struggle to balance three critical objectives: text alignment, subject identity preservation, and layout control, while the reliance on additional training further limits their scalability and efficiency. In this paper, we present AnyMS, a novel training-free framework for layout-guided multi-subject customization. AnyMS leverages three input conditions: text prompt, subject images, and layout constraints, and introduces a bottom-up dual-level attention decoupling mechanism to harmonize their integration during generation. Specifically, global decoupling separates cross-attention between textual and visual conditions to ensure text alignment. Local decoupling confines each subject's attention to its designated area, which prevents subject conflicts and thus guarantees identity preservation and layout control. Moreover, AnyMS employs pre-trained image adapters to extract subject-specific features aligned with the diffusion model, removing the need for subject learning or adapter tuning. Extensive experiments demonstrate that AnyMS achieves state-of-the-art performance, supporting complex compositions and scaling to a larger number of subjects.
Authors: Shengyi Hua, Jianfeng Wu, Tianle Shen, Kangzhe Hu, Zhongzhen Huang, Shujuan Ni, Zhihong Zhang, Yuan Li, Zhe Wang, Xiaofan Zhang
Abstract: Recent pathological foundation models have substantially advanced visual representation learning and multimodal interaction. However, most models still rely on a static inference paradigm in which whole-slide images are processed once to produce predictions, without reassessment or targeted evidence acquisition under ambiguous diagnoses. This contrasts with clinical diagnostic workflows that refine hypotheses through repeated slide observations and further examination requests. We propose PathFound, an agentic multimodal model designed to support evidence-seeking inference in pathological diagnosis. PathFound integrates the power of pathological visual foundation models, vision-language models, and reasoning models trained with reinforcement learning to perform proactive information acquisition and diagnosis refinement by progressing through the initial diagnosis, evidence-seeking, and final decision stages. Across several large multimodal models, adopting this strategy consistently improves diagnostic accuracy, indicating the effectiveness of evidence-seeking workflows in computational pathology. Among these models, PathFound achieves state-of-the-art diagnostic performance across diverse clinical scenarios and demonstrates strong potential to discover subtle details, such as nuclear features and local invasions.
Authors: Zongsheng Cao, Yangfan He, Anran Liu, Jun Xie, Feng Chen, Zepeng Wang
Abstract: Recent advances in diffusion models have notably enhanced text-to-image (T2I) generation quality, but they also raise the risk of generating unsafe content. Traditional safety methods like text blacklisting or harmful content classification have significant drawbacks: they can be easily circumvented or require extensive datasets and extra training. To overcome these challenges, we introduce PurifyGen, a novel, training-free approach for safe T2I generation that retains the model's original weights. PurifyGen introduces a dual-stage strategy for prompt purification. First, we evaluate the safety of each token in a prompt by computing its complementary semantic distance, which measures the semantic proximity between the prompt tokens and concept embeddings from predefined toxic and clean lists. This enables fine-grained prompt classification without explicit keyword matching or retraining. Tokens closer to toxic concepts are flagged as risky. Second, for risky prompts, we apply a dual-space transformation: we project toxic-aligned embeddings into the null space of the toxic concept matrix, effectively removing harmful semantic components, and simultaneously align them into the range space of clean concepts. This dual alignment purifies risky prompts by both subtracting unsafe semantics and reinforcing safe ones, while retaining the original intent and coherence. We further define a token-wise strategy to selectively replace only risky token embeddings, ensuring minimal disruption to safe content. PurifyGen offers a plug-and-play solution with theoretical grounding and strong generalization to unseen prompts and models. Extensive testing shows that PurifyGen surpasses current methods in reducing unsafe content across five datasets and competes well with training-dependent approaches. The code can refer to https://github.com/AI-Researcher-Team/PurifyGen.
Authors: Hanzheng Li, Xi Fang, Yixuan Li, Chaozheng Huang, Junjie Wang, Xi Wang, Hongzhe Bai, Bojun Hao, Shenyu Lin, Huiqi Liang, Linfeng Zhang, Guolin Ke
Abstract: The integration of Multimodal Large Language Models (MLLMs) into chemistry promises to revolutionize scientific discovery, yet their ability to comprehend the dense, graphical language of reactions within authentic literature remains underexplored. Here, we introduce RxnBench, a multi-tiered benchmark designed to rigorously evaluate MLLMs on chemical reaction understanding from scientific PDFs. RxnBench comprises two tasks: Single-Figure QA (SF-QA), which tests fine-grained visual perception and mechanistic reasoning using 1,525 questions derived from 305 curated reaction schemes, and Full-Document QA (FD-QA), which challenges models to synthesize information from 108 articles, requiring cross-modal integration of text, schemes, and tables. Our evaluation of MLLMs reveals a critical capability gap: while models excel at extracting explicit text, they struggle with deep chemical logic and precise structural recognition. Notably, models with inference-time reasoning significantly outperform standard architectures, yet none achieve 50\% accuracy on FD-QA. These findings underscore the urgent need for domain-specific visual encoders and stronger reasoning engines to advance autonomous AI chemists.
Authors: Siyu Jiao, Yiheng Lin, Yujie Zhong, Qi She, Wei Zhou, Xiaohan Lan, Zilong Huang, Fei Yu, Yingchen Yu, Yunqing Zhao, Yao Zhao, Yunchao Wei
Abstract: Recent progress in Multimodal Large Language Models (MLLMs) demonstrates that Chain-of-Thought (CoT) reasoning enables systematic solutions to complex understanding tasks. However, its extension to generation tasks remains nascent and limited by scenario-specific mechanisms that hinder generalization and adaptation. In this work, we present ThinkGen, the first think-driven visual generation framework that explicitly leverages MLLM's CoT reasoning in various generation scenarios. ThinkGen employs a decoupled architecture comprising a pretrained MLLM and a Diffusion Transformer (DiT), wherein the MLLM generates tailored instructions based on user intent, and DiT produces high-quality images guided by these instructions. We further propose a separable GRPO-based training paradigm (SepGRPO), alternating reinforcement learning between the MLLM and DiT modules. This flexible design enables joint training across diverse datasets, facilitating effective CoT reasoning for a wide range of generative scenarios. Extensive experiments demonstrate that ThinkGen achieves robust, state-of-the-art performance across multiple generation benchmarks. Code is available: https://github.com/jiaosiyuu/ThinkGen
Authors: Zhaoming Kong, Xiaowei Yang, Jiahuan Zhang
Abstract: The proliferation of imaging devices and countless image data generated every day impose an increasingly high demand on efficient and effective image denoising. In this paper, we establish a theoretical connection between principal component analysis (PCA) and the Haar transform under circulant representation, and present a computationally simple denoising algorithm. The proposed method, termed Haar-tSVD, exploits a unified tensor singular value decomposition (t-SVD) projection combined with Haar transform to efficiently capture global and local patch correlations. Haar-tSVD operates as a one-step, parallelizable plug-and-play denoiser that eliminates the need for learning local bases, thereby striking a balance between denoising speed and performance. Besides, an adaptive noise estimation scheme is introduced to improve robustness according to eigenvalue analysis of the circulant structure. To further enhance the performance under severe noise conditions, we integrate deep neural networks with Haar-tSVD based on the established Haar-PCA relationship. Experimental results on various denoising datasets demonstrate the efficiency and effectiveness of proposed method for noise removal. Our code is publicly available at https://github.com/ZhaomingKong/Haar-tSVD.
Authors: Shaohan Yu, Lijun Li, Chenyang Si, Lu Sheng, Jing Shao
Abstract: The rapid evolution of generative models has led to a continuous emergence of multimodal safety risks, exposing the limitations of existing defense methods. To address these challenges, we propose ProGuard, a vision-language proactive guard that identifies and describes out-of-distribution (OOD) safety risks without the need for model adjustments required by traditional reactive approaches. We first construct a modality-balanced dataset of 87K samples, each annotated with both binary safety labels and risk categories under a hierarchical multimodal safety taxonomy, effectively mitigating modality bias and ensuring consistent moderation across text, image, and text-image inputs. Based on this dataset, we train our vision-language base model purely through reinforcement learning (RL) to achieve efficient and concise reasoning. To approximate proactive safety scenarios in a controlled setting, we further introduce an OOD safety category inference task and augment the RL objective with a synonym-bank-based similarity reward that encourages the model to generate concise descriptions for unseen unsafe categories. Experimental results show that ProGuard achieves performance comparable to closed-source large models on binary safety classification, substantially outperforms existing open-source guard models on unsafe content categorization. Most notably, ProGuard delivers a strong proactive moderation ability, improving OOD risk detection by 52.6% and OOD risk description by 64.8%.
Authors: Ethan Chern, Zhulin Hu, Bohao Tang, Jiadi Su, Steffi Chern, Zhijie Deng, Pengfei Liu
Abstract: Real-time video generation via diffusion is essential for building general-purpose multimodal interactive AI systems. However, the simultaneous denoising of all video frames with bidirectional attention via an iterative process in diffusion models prevents real-time interaction. While existing distillation methods can make the model autoregressive and reduce sampling steps to mitigate this, they focus primarily on text-to-video generation, leaving the human-AI interaction unnatural and less efficient. This paper targets real-time interactive video diffusion conditioned on a multimodal context, including text, image, and audio, to bridge the gap. Given the observation that the leading on-policy distillation approach Self Forcing encounters challenges (visual artifacts like flickering, black frames, and quality degradation) with multimodal conditioning, we investigate an improved distillation recipe with emphasis on the quality of condition inputs as well as the initialization and schedule for the on-policy optimization. On benchmarks for multimodal-conditioned (audio, image, and text) avatar video generation including HDTF, AVSpeech, and CelebV-HQ, our distilled model matches the visual quality of the full-step, bidirectional baselines of similar or larger size with 20x less inference cost and latency. Further, we integrate our model with audio language models and long-form video inference technique Anchor-Heavy Identity Sinks to build LiveTalk, a real-time multimodal interactive avatar system. System-level evaluation on our curated multi-turn interaction benchmark shows LiveTalk outperforms state-of-the-art models (Sora2, Veo3) in multi-turn video coherence and content quality, while reducing response latency from 1 to 2 minutes to real-time generation, enabling seamless human-AI multimodal interaction.
Authors: Damiano Marsili, Aditya Mehta, Ryan Y. Lin, Georgia Gkioxari
Abstract: Vision-language models (VLMs) excel at broad visual understanding but remain coarse-grained, exhibit visual biases, and miss subtle visual details. Existing training corpora reinforce this limitation by emphasizing general recognition ("Is it a cat or a dog?") over fine-grained perception. To address this, we introduce a new training corpus and task designed to enhance the perceptual abilities of VLMs. TWIN is a large-scale dataset of 561,000 image-pair queries that task models to determine whether two visually similar images depict the same object, encouraging attention to nuanced visual cues. The dataset spans a diverse range of everyday objects across contexts, viewpoints, and appearances. Fine-tuning VLMs on TWIN yields notable gains in fine-grained recognition, even on unseen domains such as art, animals, plants, and landmarks. To quantify these gains, we introduce FGVQA, a benchmark suite of 12,000 queries that repurposes fine-grained recognition and retrieval datasets from multiple domains. While existing VLMs struggle on FGVQA, when fine-tuned on TWIN they improve by up to 19.3%, without compromising performance on general VQA benchmarks. Finally, our TWIN dataset scales favorably with object annotations, and our analysis shows that scale is key to performance. We envision TWIN as a drop-in addition to open-source VLM training corpora, advancing perceptual precision of future models. Project webpage: https://glab-caltech.github.io/twin/
Authors: Nguyen Truong Khai, Luong Duc Vinh
Abstract: Improving the accuracy of fire detection using infrared night vision cameras remains a challenging task. Previous studies have reported strong performance with popular detection models. For example, YOLOv7 achieved an mAP50-95 of 0.51 using an input image size of 640 x 1280, RT-DETR reached an mAP50-95 of 0.65 with an image size of 640 x 640, and YOLOv9 obtained an mAP50-95 of 0.598 at the same resolution. Despite these results, limitations in dataset construction continue to cause issues, particularly the frequent misclassification of bright artificial lights as fire. This report presents three main contributions: an additional NIR dataset, a two-stage detection model, and Patched-YOLO. First, to address data scarcity, we explore and apply various data augmentation strategies for both the NIR dataset and the classification dataset. Second, to improve night-time fire detection accuracy while reducing false positives caused by artificial lights, we propose a two-stage pipeline combining YOLOv11 and EfficientNetV2-B0. The proposed approach achieves higher detection accuracy compared to previous methods, particularly for night-time fire detection. Third, to improve fire detection in RGB images, especially for small and distant objects, we introduce Patched-YOLO, which enhances the model's detection capability through patch-based processing. Further details of these contributions are discussed in the following sections.
Authors: Janani Annur Thiruvengadam, Kiran Mayee Nabigaru, Anusha Kovi
Abstract: The early detection of pancreatic neoplasm is a major clinical dilemma, and it is predominantly so because tumors are likely to occur with minimal contrast margins and a large spread anatomy-wide variation amongst patients on a CT scan. These complexities require to be addressed with an effective and scalable system that can assist in enhancing the salience of the subtle visual cues and provide a high level of the generalization on the multimodal imaging data. A Scalable Residual Feature Aggregation (SRFA) framework is proposed to be used to meet these conditions in this study. The framework integrates a pipeline of preprocessing followed by the segmentation using the MAGRes-UNet that is effective in making the pancreatic structures and isolating regions of interest more visible. DenseNet-121 performed with residual feature storage is used to extract features to allow deep hierarchical features to be aggregated without properties loss. To go further, hybrid HHO-BA metaheuristic feature selection strategy is used, which guarantees the best feature subset refinement. To be classified, the system is trained based on a new hybrid model that integrates the ability to pay attention on the world, which is the Vision Transformer (ViT) with the high representational efficiency of EfficientNet-B3. A dual optimization mechanism incorporating SSA and GWO is used to fine-tune hyperparameters to enhance greater robustness and less overfitting. Experimental results support the significant improvement in performance, with the suggested model reaching 96.23% accuracy, 95.58% F1-score and 94.83% specificity, the model is significantly better than the traditional CNNs and contemporary transformer-based models. Such results highlight the possibility of the SRFA framework as a useful instrument in the early detection of pancreatic tumors.
Authors: Shu Pu, Boya Zeng, Kaichen Zhou, Mengyu Wang, Zhuang Liu
Abstract: Generative models are increasingly used in 3D vision to synthesize novel shapes, yet it remains unclear whether their generation relies on memorizing training shapes. Understanding their memorization could help prevent training data leakage and improve the diversity of generated results. In this paper, we design an evaluation framework to quantify memorization in 3D generative models and study the influence of different data and modeling designs on memorization. We first apply our framework to quantify memorization in existing methods. Next, through controlled experiments with a latent vector-set (Vecset) diffusion model, we find that, on the data side, memorization depends on data modality, and increases with data diversity and finer-grained conditioning; on the modeling side, it peaks at a moderate guidance scale and can be mitigated by longer Vecsets and simple rotation augmentation. Together, our framework and analysis provide an empirical understanding of memorization in 3D generative models and suggest simple yet effective strategies to reduce it without degrading generation quality. Our code is available at https://github.com/zlab-princeton/3d_mem.
Authors: Xiaoyu Li, Peidong Li, Xian Wu, Long Shi, Dedong Liu, Yitao Wu, Jiajia Fu, Dixiao Cui, Lijun Zhao, Lining Sun
Abstract: Spatio-temporal alignment is crucial for temporal modeling of end-to-end (E2E) perception in autonomous driving (AD), providing valuable structural and textural prior information. Existing methods typically rely on the attention mechanism to align objects across frames, simplifying the motion model with a unified explicit physical model (constant velocity, etc.). These approaches prefer semantic features for implicit alignment, challenging the importance of explicit motion modeling in the traditional perception paradigm. However, variations in motion states and object features across categories and frames render this alignment suboptimal. To address this, we propose HAT, a spatio-temporal alignment module that allows each object to adaptively decode the optimal alignment proposal from multiple hypotheses without direct supervision. Specifically, HAT first utilizes multiple explicit motion models to generate spatial anchors and motion-aware feature proposals for historical instances. It then performs multi-hypothesis decoding by incorporating semantic and motion cues embedded in cached object queries, ultimately providing the optimal alignment proposal for the target frame. On nuScenes, HAT consistently improves 3D temporal detectors and trackers across diverse baselines. It achieves state-of-the-art tracking results with 46.0% AMOTA on the test set when paired with the DETR3D detector. In an object-centric E2E AD method, HAT enhances perception accuracy (+1.3% mAP, +3.1% AMOTA) and reduces the collision rate by 32%. When semantics are corrupted (nuScenes-C), the enhancement of motion modeling by HAT enables more robust perception and planning in the E2E AD.
Authors: Keda Tao, Wenjie Du, Bohan Yu, Weiqiang Wang, Jian Liu, Huan Wang
Abstract: Omnimodal large language models have made significant strides in unifying audio and visual modalities; however, they often lack the fine-grained cross-modal understanding and have difficulty with multimodal alignment. To address these limitations, we introduce OmniAgent, a fully audio-guided active perception agent that dynamically orchestrates specialized tools to achieve more fine-grained audio-visual reasoning. Unlike previous works that rely on rigid, static workflows and dense frame-captioning, this paper demonstrates a paradigm shift from passive response generation to active multimodal inquiry. OmniAgent employs dynamic planning to autonomously orchestrate tool invocation on demand, strategically concentrating perceptual attention on task-relevant cues. Central to our approach is a novel coarse-to-fine audio-guided perception paradigm, which leverages audio cues to localize temporal events and guide subsequent reasoning. Extensive empirical evaluations on three audio-video understanding benchmarks demonstrate that OmniAgent achieves state-of-the-art performance, surpassing leading open-source and proprietary models by substantial margins of 10% - 20% accuracy.
Authors: Kang Du, Yirui Guan, Zeyu Wang
Abstract: Intrinsic image decomposition is fundamental for visual understanding, as RGB images entangle material properties, illumination, and view-dependent effects. Recent diffusion-based methods have achieved strong results for single-view intrinsic decomposition; however, extending these approaches to multi-view settings remains challenging, often leading to severe view inconsistency. We propose \textbf{Intrinsic Decomposition Transformer (IDT)}, a feed-forward framework for multi-view intrinsic image decomposition. By leveraging transformer-based attention to jointly reason over multiple input images, IDT produces view-consistent intrinsic factors in a single forward pass, without iterative generative sampling. IDT adopts a physically grounded image formation model that explicitly decomposes images into diffuse reflectance, diffuse shading, and specular shading. This structured factorization separates Lambertian and non-Lambertian light transport, enabling interpretable and controllable decomposition of material and illumination effects across views. Experiments on both synthetic and real-world datasets demonstrate that IDT achieves cleaner diffuse reflectance, more coherent diffuse shading, and better-isolated specular components, while substantially improving multi-view consistency compared to prior intrinsic decomposition methods.
Authors: Shaocong Xu, Songlin Wei, Qizhe Wei, Zheng Geng, Hong Li, Licheng Shen, Qianpu Sun, Shu Han, Bin Ma, Bohan Li, Chongjie Ye, Yuhang Zheng, Nan Wang, Saining Zhang, Hao Zhao
Abstract: Transparent objects remain notoriously hard for perception systems: refraction, reflection and transmission break the assumptions behind stereo, ToF and purely discriminative monocular depth, causing holes and temporally unstable estimates. Our key observation is that modern video diffusion models already synthesize convincing transparent phenomena, suggesting they have internalized the optical rules. We build TransPhy3D, a synthetic video corpus of transparent/reflective scenes: 11k sequences rendered with Blender/Cycles. Scenes are assembled from a curated bank of category-rich static assets and shape-rich procedural assets paired with glass/plastic/metal materials. We render RGB + depth + normals with physically based ray tracing and OptiX denoising. Starting from a large video diffusion model, we learn a video-to-video translator for depth (and normals) via lightweight LoRA adapters. During training we concatenate RGB and (noisy) depth latents in the DiT backbone and co-train on TransPhy3D and existing frame-wise synthetic datasets, yielding temporally consistent predictions for arbitrary-length input videos. The resulting model, DKT, achieves zero-shot SOTA on real and synthetic video benchmarks involving transparency: ClearPose, DREDS (CatKnown/CatNovel), and TransPhy3D-Test. It improves accuracy and temporal consistency over strong image/video baselines, and a normal variant sets the best video normal estimation results on ClearPose. A compact 1.3B version runs at ~0.17 s/frame. Integrated into a grasping stack, DKT's depth boosts success rates across translucent, reflective and diffuse surfaces, outperforming prior estimators. Together, these results support a broader claim: "Diffusion knows transparency." Generative video priors can be repurposed, efficiently and label-free, into robust, temporally coherent perception for challenging real-world manipulation.
Authors: Hau-Shiang Shiu, Chin-Yang Lin, Zhixiang Wang, Chi-Wei Hsiao, Po-Fan Yu, Yu-Chih Chen, Yu-Lun Liu
Abstract: Diffusion-based video super-resolution (VSR) methods achieve strong perceptual quality but remain impractical for latency-sensitive settings due to reliance on future frames and expensive multi-step denoising. We propose Stream-DiffVSR, a causally conditioned diffusion framework for efficient online VSR. Operating strictly on past frames, it combines a four-step distilled denoiser for fast inference, an Auto-regressive Temporal Guidance (ARTG) module that injects motion-aligned cues during latent denoising, and a lightweight temporal-aware decoder with a Temporal Processor Module (TPM) that enhances detail and temporal coherence. Stream-DiffVSR processes 720p frames in 0.328 seconds on an RTX4090 GPU and significantly outperforms prior diffusion-based methods. Compared with the online SOTA TMP, it boosts perceptual quality (LPIPS +0.095) while reducing latency by over 130x. Stream-DiffVSR achieves the lowest latency reported for diffusion-based VSR, reducing initial delay from over 4600 seconds to 0.328 seconds, thereby making it the first diffusion VSR method suitable for low-latency online deployment. Project page: https://jamichss.github.io/stream-diffvsr-project-page/
URLs: https://jamichss.github.io/stream-diffvsr-project-page/
Authors: Mahadev Sunil Kumar, Arnab Raha, Debayan Das, Gopakumar G, Amitava Mukherjee
Abstract: Deep distributed networks (DNNs) have become central to modern computer vision, yet their deployment on resource-constrained edge devices remains hindered by substantial parameter counts and computational demands. Here, we present an approach to the efficient deployment of distributed DNNs that jointly respects hardware limitations and preserves task performance. Our method integrates a structured model pruning with a multi-objective optimization to tailor network capacity to heterogeneous device constraints. We demonstrate this framework using Multi-View Convolutional Neural Network (MVCNN), a state-of-the-art architecture for 3D object recognition, by quantifying the contribution of individual views to classification accuracy and allocating pruning budgets, respectively. Experimental results show that the resulting models satisfy user-specified bounds on accuracy and memory footprint while reducing inference latency by factors ranging from 1.2x to 5.0x across diverse hardware platforms. These findings suggest that performance-aware, view-adaptive compression provides a viable pathway for deploying complex vision models in distributed edge environments.
Authors: Jiesong Lian, Ruizhe Zhong, Zixiang Zhou, Xiaoyue Mi, Yixue Hao, Yuan Zhou, Qinglin Lu, Long Hu, Junchi Yan
Abstract: Post-training alignment of video generation models with human preferences is a critical goal. Developing effective Reward Models (RMs) for this process faces significant methodological hurdles. Current data collection paradigms, reliant on in-prompt pairwise annotations, suffer from labeling noise. Concurrently, the architectural design of VLM-based RMs, particularly their output mechanisms, remains underexplored. Furthermore, RM is susceptible to reward hacking in post-training. To mitigate these limitations, we propose SoliReward, a systematic framework for video RM training. Our framework first sources high-quality, cost-efficient data via single-item binary annotations, then constructs preference pairs using a cross-prompt pairing strategy. Architecturally, we employ a Hierarchical Progressive Query Attention mechanism to enhance feature aggregation. Finally, we introduce a modified BT loss that explicitly accommodates win-tie scenarios. This approach regularizes the RM's score distribution for positive samples, providing more nuanced preference signals to alleviate over-focus on a small number of top-scoring samples. Our approach is validated on benchmarks evaluating physical plausibility, subject deformity, and semantic alignment, demonstrating improvements in direct RM evaluation metrics and in the efficacy of post-training on video generation models. Code and benchmark will be publicly available.
Authors: Muhammad Ibtsaam Qadir, Duane Schonlau, Ulrike Dydak, Fiona R. Kolbinger
Abstract: This study quantitatively evaluates the impact of MRI scanner magnetic field strength on the performance and generalizability of deep learning-based segmentation algorithms. Three publicly available MRI datasets (breast tumor, pancreas, and cervical spine) were stratified by scanner field strength (1.5T vs. 3.0T). For each segmentation task, three nnU-Net-based models were developed: A model trained on 1.5T data only (m-1.5T), a model trained on 3.0T data only (m-3.0T), and a model trained on pooled 1.5T and 3.0T data (m-combined). Each model was evaluated on both 1.5T and 3.0T validation sets. Field-strength-dependent performance differences were investigated via Uniform Manifold Approximation and Projection (UMAP)-based clustering and radiomic analysis, including 23 first-order and texture features. For breast tumor segmentation, m-3.0T (DSC: 0.494 [1.5T] and 0.433 [3.0T]) significantly outperformed m-1.5T (DSC: 0.411 [1.5T] and 0.289 [3.0T]) and m-combined (DSC: 0.373 [1.5T] and 0.268[3.0T]) on both validation sets (p<0.0001). Pancreas segmentation showed similar trends: m-3.0T achieved the highest DSC (0.774 [1.5T], 0.840 [3.0T]), while m-1.5T underperformed significantly (p<0.0001). For cervical spine, models performed optimally on same-field validation sets with minimal cross-field performance degradation (DSC>0.92 for all comparisons). Radiomic analysis revealed moderate field-strength-dependent clustering in soft tissues (silhouette scores 0.23-0.29) but minimal separation in osseous structures (0.12). These results indicate that magnetic field strength in the training data substantially influences the performance of deep learning-based segmentation models, particularly for soft-tissue structures (e.g., small lesions). This warrants consideration of magnetic field strength as a confounding factor in studies evaluating AI performance on MRI.
Authors: Areeb Ehsan
Abstract: Timely brain tumor diagnosis remains challenging in low-resource clinical environments where expert neuroradiology interpretation, high-end MRI hardware, and invasive biopsy procedures may be limited. Although deep learning has achieved strong performance in brain tumor analysis, real-world adoption is constrained by computational demands, dataset shift across scanners, and limited interpretability. This paper presents a prototype virtual biopsy pipeline for four-class classification of 2D brain MRI images using a lightweight convolutional neural network (CNN) and complementary radiomics-style handcrafted features. A MobileNetV2-based CNN is trained for classification, while an interpretable radiomics branch extracts eight features capturing lesion shape, intensity statistics, and gray-level co-occurrence matrix (GLCM) texture descriptors. A late fusion strategy concatenates CNN embeddings with radiomics features and trains a RandomForest classifier on the fused representation. Explainability is provided via Grad-CAM visualizations and radiomics feature importance analysis. Experiments on a public Kaggle brain tumor MRI dataset show improved validation performance for fusion relative to single-branch baselines, while robustness tests under reduced resolution and additive noise highlight sensitivity relevant to low-resource imaging conditions. The system is framed as decision support and not a substitute for clinical diagnosis or histopathology.
Authors: Muhammad Usman, Sung-Min Gho
Abstract: Susceptibility Map Weighted Imaging (SMWI) is an advanced magnetic resonance imaging technique used to detect nigral hyperintensity in Parkinsons disease. However, full resolution SMWI acquisition is limited by long scan times. Efficient reconstruction methods are therefore required to generate high quality SMWI from reduced k space data while preserving diagnostic relevance. In this work, we propose a complex valued Swin Transformer based network for super resolution reconstruction of multi echo MRI data. The proposed method reconstructs high quality SMWI images from low resolution k space inputs. Experimental results demonstrate that the method achieves a structural similarity index of 0.9116 and a mean squared error of 0.076 when reconstructing SMWI from 256 by 256 k space data, while maintaining critical diagnostic features. This approach enables high quality SMWI reconstruction from reduced k space sampling, leading to shorter scan times without compromising diagnostic detail. The proposed method has the potential to improve the clinical applicability of SMWI for Parkinsons disease and support faster and more efficient neuroimaging workflows.
Authors: Pu Zhao, Xuan Shen, Zhenglun Kong, Yixin Shen, Sung-En Chang, Arash Akbari, Timothy Rupprecht, Lei Lu, Enfu Nan, Changdi Yang, Yumei He, Weiyan Shi, Xingchen Xu, Yu Huang, Wei Jiang, Wei Wang, Yue Chen, Yong He, Yanzhi Wang
Abstract: Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA and Mistral, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Moxin 7B is introduced as a fully open-source LLM developed in accordance with the Model Openness Framework, which moves beyond the simple sharing of model weights to embrace complete transparency in training, datasets, and implementation detail, thus fostering a more inclusive and collaborative research environment that can sustain a healthy open-source ecosystem. To further equip Moxin with various capabilities in different tasks, we develop three variants based on Moxin, including Moxin-VLM, Moxin-VLA, and Moxin-Chinese, which target the vision-language, vision-language-action, and Chinese capabilities, respectively. Experiments show that our models achieve superior performance in various evaluations. We adopt open-source framework and open data for the training. We release our models, along with the available data and code to derive these models.
Authors: Haozhe Jia
Abstract: Capsule endoscopy has enabled minimally invasive gastrointestinal imaging, but its clinical utility is limited by the inherently low resolution of captured images due to hardware, power, and transmission constraints. This limitation hampers the identification of fine-grained mucosal textures and subtle pathological features essential for early diagnosis. This work investigates a diffusion-based super-resolution framework to enhance capsule endoscopy images in a data-driven and anatomically consistent manner. We adopt the SR3 (Super-Resolution via Repeated Refinement) framework built upon Denoising Diffusion Probabilistic Models (DDPMs) to learn a probabilistic mapping from low-resolution to high-resolution images. Unlike GAN-based approaches that often suffer from training instability and hallucination artifacts, diffusion models provide stable likelihood-based training and improved structural fidelity. The HyperKvasir dataset, a large-scale publicly available gastrointestinal endoscopy dataset, is used for training and evaluation. Quantitative results demonstrate that the proposed method significantly outperforms bicubic interpolation and GAN-based super-resolution methods such as ESRGAN, achieving PSNR of 27.5 dB and SSIM of 0.65 for a baseline model, and improving to 29.3 dB and 0.71 with architectural enhancements including attention mechanisms. Qualitative results show improved preservation of anatomical boundaries, vascular patterns, and lesion structures. These findings indicate that diffusion-based super-resolution is a promising approach for enhancing non-invasive medical imaging, particularly in capsule endoscopy where image resolution is fundamentally constrained.
Authors: Byung-Kwan Lee, Yu-Chiang Frank Wang, Ryo Hachiuma
Abstract: Large-scale vision-language models (VLMs) have recently achieved remarkable multimodal understanding, but their massive size makes them impractical for deployment on mobile or edge devices. This raises the need for compact yet capable VLMs that can efficiently learn from powerful large teachers. However, distilling knowledge from a large teacher to a small student remains challenging due to their large size gap: the student often fails to reproduce the teacher's complex, high-dimensional representations, leading to unstable learning and degraded performance. To address this, we propose Masters (Masking Teacher and Reinforcing Student), a mask-progressive reinforcement learning (RL) distillation framework. Masters first masks non-dominant weights of the teacher to reduce unnecessary complexity, then progressively restores the teacher by gradually increasing its capacity during training. This strategy allows the student to learn richer representations from the teacher in a smooth and stable manner. To further refine knowledge transfer, Masters integrates an offline RL stage with two complementary rewards: an accuracy reward that measures the correctness of the generated responses, and a distillation reward that quantifies the ease of transferring responses from teacher to student. Unlike online think-answer RL paradigms that are computationally expensive and generate lengthy responses, our offline RL leverages pre-generated responses from masked teachers. These provide rich yet efficient guidance, enabling students to achieve strong performance without requiring the think-answer process.
Authors: Shaurya Gaur, Michel Vitale, Alessa Hering, Johan Kwisthout, Colin Jacobs, Lena Philipp, Fennie van der Graaf
Abstract: Lung cancer is the leading cause of cancer-related mortality in adults worldwide. Screening high-risk individuals with annual low-dose CT (LDCT) can support earlier detection and reduce deaths, but widespread implementation may strain the already limited radiology workforce. AI models have shown potential in estimating lung cancer risk from LDCT scans. However, high-risk populations for lung cancer are diverse, and these models' performance across demographic groups remains an open question. In this study, we drew on the considerations on confounding factors and ethically significant biases outlined in the JustEFAB framework to evaluate potential performance disparities and fairness in two deep learning risk estimation models for lung cancer screening: the Sybil lung cancer risk model and the Venkadesh21 nodule risk estimator. We also examined disparities in the PanCan2b logistic regression model recommended in the British Thoracic Society nodule management guideline. Both deep learning models were trained on data from the US-based National Lung Screening Trial (NLST), and assessed on a held-out NLST validation set. We evaluated AUROC, sensitivity, and specificity across demographic subgroups, and explored potential confounding from clinical risk factors. We observed a statistically significant AUROC difference in Sybil's performance between women (0.88, 95% CI: 0.86, 0.90) and men (0.81, 95% CI: 0.78, 0.84, p < .001). At 90% specificity, Venkadesh21 showed lower sensitivity for Black (0.39, 95% CI: 0.23, 0.59) than White participants (0.69, 95% CI: 0.65, 0.73). These differences were not explained by available clinical confounders and thus may be classified as unfair biases according to JustEFAB. Our findings highlight the importance of improving and monitoring model performance across underrepresented subgroups, and further research on algorithmic fairness, in lung cancer screening.
Authors: Zheng Xing, Weibing Zhao
Abstract: Unsupervised human motion segmentation (HMS) can be effectively achieved using subspace clustering techniques. However, traditional methods overlook the role of temporal semantic exploration in HMS. This paper explores the use of temporal vision semantics (TVS) derived from human motion sequences, leveraging the image-to-text capabilities of a large language model (LLM) to enhance subspace clustering performance. The core idea is to extract textual motion information from consecutive frames via LLM and incorporate this learned information into the subspace clustering framework. The primary challenge lies in learning TVS from human motion sequences using LLM and integrating this information into subspace clustering. To address this, we determine whether consecutive frames depict the same motion by querying the LLM and subsequently learn temporal neighboring information based on its response. We then develop a TVS-integrated subspace clustering approach, incorporating subspace embedding with a temporal regularizer that induces each frame to share similar subspace embeddings with its temporal neighbors. Additionally, segmentation is performed based on subspace embedding with a temporal constraint that induces the grouping of each frame with its temporal neighbors. We also introduce a feedback-enabled framework that continuously optimizes subspace embedding based on the segmentation output. Experimental results demonstrate that the proposed method outperforms existing state-of-the-art approaches on four benchmark human motion datasets.
Authors: Renping Zhou, Zanlin Ni, Tianyi Chen, Zeyu Liu, Yang Yue, Yulin Wang, Yuxuan Wang, Jingshu Liu, Gao Huang
Abstract: Recently, Masked Diffusion Models (MDMs) have shown promising potential across vision, language, and cross-modal generation. However, a notable discrepancy exists between their training and inference procedures. In particular, MDM inference is a multi-step, iterative process governed not only by the model itself but also by various schedules that dictate the token-decoding trajectory (e.g., how many tokens to decode at each step). In contrast, MDMs are typically trained using a simplified, single-step BERT-style objective that masks a subset of tokens and predicts all of them simultaneously. This step-level simplification fundamentally disconnects the training paradigm from the trajectory-level nature of inference, leaving the inference schedules never optimized during training. In this paper, we introduce Co-GRPO, which reformulates MDM generation as a unified Markov Decision Process (MDP) that jointly incorporates both the model and the inference schedule. By applying Group Relative Policy Optimization at the trajectory level, Co-GRPO cooperatively optimizes model parameters and schedule parameters under a shared reward, without requiring costly backpropagation through the multi-step generation process. This holistic optimization aligns training with inference more thoroughly and substantially improves generation quality. Empirical results across four benchmarks-ImageReward, HPS, GenEval, and DPG-Bench-demonstrate the effectiveness of our approach. For more details, please refer to our project page: https://co-grpo.github.io/ .
Authors: Xudong Ling, Tianxi Huang, Qian Dong, Tao He, Chaorong Li, Guiduo Duan
Abstract: Short-term precipitation nowcasting is an inherently uncertain and under-constrained spatiotemporal forecasting problem, especially for rapidly evolving and extreme weather events. Existing generative approaches rely primarily on visual conditioning, leaving future motion weakly constrained and ambiguous. We propose a language-aware multimodal nowcasting framework(LangPrecip) that treats meteorological text as a semantic motion constraint on precipitation evolution. By formulating nowcasting as a semantically constrained trajectory generation problem under the Rectified Flow paradigm, our method enables efficient and physically consistent integration of textual and radar information in latent space.We further introduce LangPrecip-160k, a large-scale multimodal dataset with 160k paired radar sequences and motion descriptions. Experiments on Swedish and MRMS datasets show consistent improvements over state-of-the-art methods, achieving over 60 \% and 19\% gains in heavy-rainfall CSI at an 80-minute lead time.
Authors: Shaofei Cai, Yulei Qin, Haojia Lin, Zihan Xu, Gang Li, Yuchen Shi, Zongyi Li, Yong Mao, Siqi Cai, Xiaoyu Tan, Yitao Liang, Ke Li, Xing Sun
Abstract: Agentic reinforcement learning (RL) holds great promise for the development of autonomous agents under complex GUI tasks, but its scalability remains severely hampered by the verification of task completion. Existing task verification is treated as a passive, post-hoc process: a verifier (i.e., rule-based scoring script, reward or critic model, and LLM-as-a-Judge) analyzes the agent's entire interaction trajectory to determine if the agent succeeds. Such processing of verbose context that contains irrelevant, noisy history poses challenges to the verification protocols and therefore leads to prohibitive cost and low reliability. To overcome this bottleneck, we propose SmartSnap, a paradigm shift from this passive, post-hoc verification to proactive, in-situ self-verification by the agent itself. We introduce the Self-Verifying Agent, a new type of agent designed with dual missions: to not only complete a task but also to prove its accomplishment with curated snapshot evidences. Guided by our proposed 3C Principles (Completeness, Conciseness, and Creativity), the agent leverages its accessibility to the online environment to perform self-verification on a minimal, decisive set of snapshots. Such evidences are provided as the sole materials for a general LLM-as-a-Judge verifier to determine their validity and relevance. Experiments on mobile tasks across model families and scales demonstrate that our SmartSnap paradigm allows training LLM-driven agents in a scalable manner, bringing performance gains up to 26.08% and 16.66% respectively to 8B and 30B models. The synergizing between solution finding and evidence seeking facilitates the cultivation of efficient, self-verifying agents with competitive performance against DeepSeek V3.1 and Qwen3-235B-A22B.
Authors: Elsen Ronando, Sozo Inoue
Abstract: In this paper, we propose an LLM-Guided Exemplar Selection framework to address a key limitation in state-of-the-art Human Activity Recognition (HAR) methods: their reliance on large labeled datasets and purely geometric exemplar selection, which often fail to distinguish similar weara-ble sensor activities such as walking, walking upstairs, and walking downstairs. Our method incorporates semantic reasoning via an LLM-generated knowledge prior that captures feature importance, inter-class confusability, and exemplar budget multipliers, and uses it to guide exemplar scoring and selection. These priors are combined with margin-based validation cues, PageRank centrality, hubness penalization, and facility-location optimization to obtain a compact and informative set of exemplars. Evaluated on the UCI-HAR dataset under strict few-shot conditions, the framework achieves a macro F1-score of 88.78%, outperforming classical approaches such as random sampling, herding, and $k$-center. The results show that LLM-derived semantic priors, when integrated with structural and geometric cues, provide a stronger foundation for selecting representative sensor exemplars in few-shot wearable-sensor HAR.
Authors: Kai-Hsiang Hsieh, Monyneath Yim, Wen-Hsiao Peng, Jui-Chiu Chiang
Abstract: Joint compression of point cloud geometry and attributes is essential for efficient 3D data representation. Existing methods often rely on post-hoc recoloring procedures and manually tuned bitrate allocation between geometry and attribute bitstreams in inference, which hinders end-to-end optimization and increases system complexity. To overcome these limitations, we propose MEGA-PCC, a fully end-to-end, learning-based framework featuring two specialized models for joint compression. The main compression model employs a shared encoder that encodes both geometry and attribute information into a unified latent representation, followed by dual decoders that sequentially reconstruct geometry and then attributes. Complementing this, the Mamba-based Entropy Model (MEM) enhances entropy coding by capturing spatial and channel-wise correlations to improve probability estimation. Both models are built on the Mamba architecture to effectively model long-range dependencies and rich contextual features. By eliminating the need for recoloring and heuristic bitrate tuning, MEGA-PCC enables data-driven bitrate allocation during training and simplifies the overall pipeline. Extensive experiments demonstrate that MEGA-PCC achieves superior rate-distortion performance and runtime efficiency compared to both traditional and learning-based baselines, offering a powerful solution for AI-driven point cloud compression.
Authors: Jian Li, Karthik Gopinath, Brian L. Edlow, Adrian V. Dalca, Bruce Fischl
Abstract: Cortical surface parcellation is a fundamental task in both basic neuroscience research and clinical applications, enabling more accurate mapping of brain regions. Model-based and learning-based approaches for automated parcellation alleviate the need for manual labeling. Despite the advancement in parcellation performance, learning-based methods shift away from registration and atlas propagation without exploring the reason for the improvement compared to traditional methods. In this study, we present JParc, a joint cortical registration and parcellation framework, that outperforms existing state-of-the-art parcellation methods. In rigorous experiments, we demonstrate that the enhanced performance of JParc is primarily attributable to accurate cortical registration and a learned parcellation atlas. By leveraging a shallow subnetwork to fine-tune the propagated atlas labels, JParc achieves a Dice score greater than 90% on the Mindboggle dataset, using only basic geometric features (sulcal depth, curvature) that describe cortical folding patterns. The superior accuracy of JParc can significantly increase the statistical power in brain mapping studies as well as support applications in surgical planning and many other downstream neuroscientific and clinical tasks.
Authors: Hassan Wasswa, Timothy Lynar
Abstract: Although AI-based models have achieved high accuracy in IoT threat detection, their deployment in enterprise environments is constrained by reliance on stationary datasets that fail to reflect the dynamic nature of real-world IoT NetFlow traffic, which is frequently affected by concept drift. Existing solutions typically rely on periodic classifier retraining, resulting in high computational overhead and the risk of catastrophic forgetting. To address these challenges, this paper proposes a scalable framework for adaptive IoT threat detection that eliminates the need for continuous classifier retraining. The proposed approach trains a classifier once on latent-space representations of historical traffic, while an alignment model maps incoming traffic to the learned historical latent space prior to classification, thereby preserving knowledge of previously observed attacks. To capture inter-instance relationships among attack samples, the low-dimensional latent representations are further transformed into a graph-structured format and classified using a graph neural network. Experimental evaluations on real-world heterogeneous IoT traffic datasets demonstrate that the proposed framework maintains robust detection performance under concept drift. These results highlight the framework's potential for practical deployment in dynamic and large-scale IoT environments.
Authors: Borong Zhang, Jiahao Li, Jiachen Shen, Yishuai Cai, Yuhao Zhang, Yuanpei Chen, Juntao Dai, Jiaming Ji, Yaodong Yang
Abstract: While Vision-Language-Action models (VLAs) are rapidly advancing towards generalist robot policies, it remains difficult to quantitatively understand their limits and failure modes. To address this, we introduce a comprehensive benchmark called VLA-Arena. We propose a novel structured task design framework to quantify difficulty across three orthogonal axes: (1) Task Structure, (2) Language Command, and (3) Visual Observation. This allows us to systematically design tasks with fine-grained difficulty levels, enabling a precise measurement of model capability frontiers. For Task Structure, VLA-Arena's 170 tasks are grouped into four dimensions: Safety, Distractor, Extrapolation, and Long Horizon. Each task is designed with three difficulty levels (L0-L2), with fine-tuning performed exclusively on L0 to assess general capability. Orthogonal to this, language (W0-W4) and visual (V0-V4) perturbations can be applied to any task to enable a decoupled analysis of robustness. Our extensive evaluation of state-of-the-art VLAs reveals several critical limitations, including a strong tendency toward memorization over generalization, asymmetric robustness, a lack of consideration for safety constraints, and an inability to compose learned skills for long-horizon tasks. To foster research addressing these challenges and ensure reproducibility, we provide the complete VLA-Arena framework, including an end-to-end toolchain from task definition to automated evaluation and the VLA-Arena-S/M/L datasets for fine-tuning. Our benchmark, data, models, and leaderboard are available at https://vla-arena.github.io.
Authors: Junshu Dai, Yu Wang, Tongya Zheng, Wei Ji, Qinghong Guo, Ji Cao, Jie Song, Canghong Jin, Mingli Song
Abstract: The precise prediction of human mobility has produced significant socioeconomic impacts, such as location recommendations and evacuation suggestions. However, existing methods suffer from limited generalization capability: unimodal approaches are constrained by data sparsity and inherent biases, while multi-modal methods struggle to effectively capture mobility dynamics caused by the semantic gap between static multi-modal representation and spatial-temporal dynamics. Therefore, we leverage multi-modal spatial-temporal knowledge to characterize mobility dynamics for the location recommendation task, dubbed as \textbf{M}ulti-\textbf{M}odal \textbf{Mob}ility (\textbf{M}$^3$\textbf{ob}). First, we construct a unified spatial-temporal relational graph (STRG) for multi-modal representation, by leveraging the functional semantics and spatial-temporal knowledge captured by the large language models (LLMs)-enhanced spatial-temporal knowledge graph (STKG). Second, we design a gating mechanism to fuse spatial-temporal graph representations of different modalities, and propose an STKG-guided cross-modal alignment to inject spatial-temporal dynamic knowledge into the static image modality. Extensive experiments on six public datasets show that our proposed method not only achieves consistent improvements in normal scenarios but also exhibits significant generalization ability in abnormal scenarios.
Authors: Jiashu Dong, Jiabing Xiang, Lisheng Geng, Suqing Tian, Wei Zhao
Abstract: X-ray computed tomography (CT) is widely used in medical imaging, with sparse-view reconstruction offering an effective way to reduce radiation dose. However, ill-posed conditions often result in severe streak artifacts. Recent advances in deep learning-based methods have improved reconstruction quality, but challenges still remain. To address these challenges, we propose a novel semantic feature contrastive learning loss function that evaluates semantic similarity in high-level latent spaces and anatomical similarity in shallow latent spaces. Our approach utilizes a three-stage U-Net-based architecture: one for coarse reconstruction, one for detail refinement, and one for semantic similarity measurement. Tests on a chest dataset with orthogonal projections demonstrate that our method achieves superior reconstruction quality and faster processing compared to other algorithms. The results show significant improvements in image quality while maintaining low computational complexity, making it a practical solution for orthogonal CT reconstruction.
Authors: Poojan Vanani, Darsh Patel, Danyal Khorami, Siva Munaganuru, Pavan Reddy, Varun Reddy, Bhargav Raghunath, Ishrat Lallmamode, Romir Patel, Assegid Kidan\'e, Tejaswi Gowda
Abstract: Motion capture remains costly and complex to deploy, limiting use outside specialized laboratories. We present Mesquite, an open-source, low-cost inertial motion-capture system that combines a body-worn network of 15 IMU sensor nodes with a hip-worn Android smartphone for position tracking. A low-power wireless link streams quaternion orientations to a central USB dongle and a browser-based application for real-time visualization and recording. Built on modern web technologies -- WebGL for rendering, WebXR for SLAM, WebSerial and WebSockets for device and network I/O, and Progressive Web Apps for packaging -- the system runs cross-platform entirely in the browser. In benchmarks against a commercial optical system, Mesquite achieves mean joint-angle error of 2-5 degrees while operating at approximately 5% of the cost. The system sustains 30 frames per second with end-to-end latency under 15ms and a packet delivery rate of at least 99.7% in standard indoor environments. By leveraging IoT principles, edge processing, and a web-native stack, Mesquite lowers the barrier to motion capture for applications in entertainment, biomechanics, healthcare monitoring, human-computer interaction, and virtual reality. We release hardware designs, firmware, and software under an open-source license (GNU GPL).
Authors: Jun Wang
Abstract: We propose a theoretical framework for continual and experiential learning in large language model agents that integrates episodic memory with reinforcement learning. The framework identifies reflection as the key mechanism that enables agents to adapt through interaction without back propagation or model fine tuning, thereby relaxing the conventional separation between training and deployment.To formalise this process, we introduce the Stateful Reflective Decision Process, which models reflective learning as a two stage read write interaction with episodic memory. Writing stores interaction outcomes and corresponds to policy evaluation, while reading retrieves relevant past cases and corresponds to policy improvement. We show that this process induces an equivalent Markov decision process over augmented state memory representations, allowing the use of classical tools from dynamic programming and reinforcement learning. We further instantiate the framework using entropy regularised policy iteration and establish convergence guarantees. As episodic memory grows and achieves sufficient coverage of the state space, the resulting policy converges to the optimal solution. This work provides a principled foundation for memory augmented and retrieval based language model agents capable of continual adaptation without parameter updates.
Authors: Minghao Dong, Xinyang Luo, Xujian Ouyang, Yongshun Xiao
Abstract: Compton cameras (CCs) are a kind of gamma cameras which are designed to determine the directions of incident gammas based on the Compton scatter. However, the reconstruction of CCs face problems of severe artifacts and deformation due to the fundamental reconstruction principle of back-projection of Compton cones. Besides, a part of systematic errors originated from the performance of devices are hard to remove through calibration, leading to deterioration of imaging quality. Iterative algorithms and deep-learning based methods have been widely used to improve reconstruction. But most of them are optimization based on the results of back-projection. Therefore, we proposed an end-to-end deep learning framework, SwinCCIR, for CC imaging. Through adopting swin-transformer blocks and a transposed convolution-based image generation module, we established the relationship between the list-mode events and the radioactive source distribution. SwinCCIR was trained and validated on both simulated and practical dataset. The experimental results indicate that SwinCCIR effectively overcomes problems of conventional CC imaging, which are expected to be implemented in practical applications.
Authors: Truong Son Nguyen
Abstract: We introduce \textbf{Schr\"{o}dinger AI}, a unified machine learning framework inspired by quantum mechanics. The system is defined by three tightly coupled components: (1) a {time-independent wave-energy solver} that treats perception and classification as spectral decomposition under a learned Hamiltonian; (2) a {time-dependent dynamical solver} governing the evolution of semantic wavefunctions over time, enabling context-aware decision revision, re-routing, and reasoning under environmental changes; and (3) a {low-rank operator calculus} that learns symbolic transformations such as modular arithmetic through learned quantum-like transition operators. Together, these components form a coherent physics-driven alternative to conventional cross-entropy training and transformer attention, providing robust generalization, interpretable semantics, and emergent topology. Empirically, Schr\"{o}dinger AI demonstrates: (a) emergent semantic manifolds that reflect human-conceived class relations without explicit supervision; (b) dynamic reasoning that adapts to changing environments, including maze navigation with real-time potential-field perturbations; and (c) exact operator generalization on modular arithmetic tasks, where the system learns group actions and composes them across sequences far beyond training length. These results suggest a new foundational direction for machine learning, where learning is cast as discovering and navigating an underlying semantic energy landscape.
Authors: Amirhossein Tighkhorshid, Zahra Dehghanian, Gholamali Aminian, Chengchun Shi, Hamid R. Rabiee
Abstract: Distillation addresses the slow sampling problem in diffusion models by creating models with smaller size or fewer steps that approximate the behavior of high-step teachers. In this work, we propose a reinforcement learning based distillation framework for diffusion models. Instead of relying on fixed reconstruction or consistency losses, we treat the distillation process as a policy optimization problem, where the student is trained using a reward signal derived from alignment with the teacher's outputs. This RL driven approach dynamically guides the student to explore multiple denoising paths, allowing it to take longer, optimized steps toward high-probability regions of the data distribution, rather than relying on incremental refinements. Our framework utilizes the inherent ability of diffusion models to handle larger steps and effectively manage the generative process. Experimental results show that our method achieves superior performance with significantly fewer inference steps and computational resources compared to existing distillation techniques. Additionally, the framework is model agnostic, applicable to any type of diffusion models with suitable reward functions, providing a general optimization paradigm for efficient diffusion learning.
Authors: Alexandru Hegyi
Abstract: Consistent glacier boundary delineation is essential for monitoring glacier change, yet many existing approaches are difficult to scale across long time series and heterogeneous environments. In this report, we present a GeoSAM-based, semi-automatic workflow for rapid glacier delineation from Sentinel-2 surface reflectance imagery. The method combines late-summer image compositing, spectral-index-based identification of candidate ice areas, prompt-guided segmentation using GeoSAM, and physically based post-processing to derive annual glacier outlines. The workflow is demonstrated in the Ny-Alesund and Kongsfjorden region of western Svalbard across multiple years of the Sentinel-2 era. Results show that the approach produces spatially coherent and temporally consistent outlines for major glacier bodies, while most errors are associated with small features affected by water bodies, terrain shadows, or high surface variability. The reliance on derived RGB imagery makes the method flexible and transferable to other optical datasets, with improved performance expected at higher spatial resolution. Although user inspection remains necessary to filter incorrect polygons and adjust thresholds for local conditions, the workflow provides a fast and practical alternative for multi-temporal glacier mapping and ice-loss assessment.
Authors: Yaping Zhang, Qixuan Zhang, Xingquan Zhang, Zhiyuan Chen, Wenwen Zhuang, Yupu Liang, Lu Xiang, Yang Zhao, Jiajun Zhang, Yu Zhou, Chengqing Zong
Abstract: The rapid advancement of large language models (LLMs) and multimodal foundation models has sparked growing interest in their potential for scientific research. However, scientific intelligence encompasses a broad spectrum of abilities ranging from understanding fundamental knowledge to conducting creative discovery, and existing benchmarks remain fragmented. Most focus on narrow tasks and fail to reflect the hierarchical and multi-disciplinary nature of real scientific inquiry. We introduce \textbf{HiSciBench}, a hierarchical benchmark designed to evaluate foundation models across five levels that mirror the complete scientific workflow: \textit{Scientific Literacy} (L1), \textit{Literature Parsing} (L2), \textit{Literature-based Question Answering} (L3), \textit{Literature Review Generation} (L4), and \textit{Scientific Discovery} (L5). HiSciBench contains 8,735 carefully curated instances spanning six major scientific disciplines, including mathematics, physics, chemistry, biology, geography, and astronomy, and supports multimodal inputs including text, equations, figures, and tables, as well as cross-lingual evaluation. Unlike prior benchmarks that assess isolated abilities, HiSciBench provides an integrated, dependency-aware framework that enables detailed diagnosis of model capabilities across different stages of scientific reasoning. Comprehensive evaluations of leading models, including GPT-5, DeepSeek-R1, and several multimodal systems, reveal substantial performance gaps: while models achieve up to 69\% accuracy on basic literacy tasks, performance declines sharply to 25\% on discovery-level challenges. HiSciBench establishes a new standard for evaluating scientific Intelligence and offers actionable insights for developing models that are not only more capable but also more reliable. The benchmark will be publicly released to facilitate future research.
Authors: Fuyad Hasan Bhoyan, Prashanta Sarker, Parsia Noor Ethila, Md. Emon Hossain, Md Kaviul Hossain, Md Humaion Kabir Mehedi
Abstract: Early and accurate detection of gallbladder diseases is crucial, yet ultrasound interpretation is challenging. To address this, an AI-driven diagnostic software integrates our hybrid deep learning model MobResTaNet to classify ten categories, nine gallbladder disease types and normal directly from ultrasound images. The system delivers interpretable, real-time predictions via Explainable AI (XAI) visualizations, supporting transparent clinical decision-making. It achieves up to 99.85% accuracy with only 2.24M parameters. Deployed as web and mobile applications using HTML, CSS, JavaScript, Bootstrap, and Flutter, the software provides efficient, accessible, and trustworthy diagnostic support at the point of care
Authors: Mingyuan Zhang, Yue Bai, Yifan Wang, Yiyang Huang, Yun Fu
Abstract: Explorations in fine-tuning Vision-Language Models (VLMs), such as Low-Rank Adaptation (LoRA) from Parameter Efficient Fine-Tuning (PEFT), have made impressive progress. However, most approaches rely on explicit weight updates, overlooking the extensive representational structures already encoded in pre-trained models that remain underutilized. Recent works have demonstrated that Mask Fine-Tuning (MFT) can be a powerful and efficient post-training paradigm for language models. Instead of updating weights, MFT assigns learnable gating scores to each weight, allowing the model to reorganize its internal subnetworks for downstream task adaptation. In this paper, we rethink fine-tuning for VLMs from a structural reparameterization perspective grounded in MFT. We apply MFT to the language and projector components of VLMs with different language backbones and compare against strong PEFT baselines. Experiments show that MFT consistently surpasses LoRA variants and even full fine-tuning, achieving high performance without altering the frozen backbone. Our findings reveal that effective adaptation can emerge not only from updating weights but also from reestablishing connections among the model's existing knowledge. Code available at: https://github.com/Ming-K9/MFT-VLM
Authors: Jianping Mei, Michael Moses, Jan Waelty, Yucheng Yang
Abstract: We study how deep learning can improve valuation in the art market by incorporating the visual content of artworks into predictive models. Using a large repeated-sales dataset from major auction houses, we benchmark classical hedonic regressions and tree-based methods against modern deep architectures, including multi-modal models that fuse tabular and image data. We find that while artist identity and prior transaction history dominate overall predictive power, visual embeddings provide a distinct and economically meaningful contribution for fresh-to-market works where historical anchors are absent. Interpretability analyses using Grad-CAM and embedding visualizations show that models attend to compositional and stylistic cues. Our findings demonstrate that multi-modal deep learning delivers significant value precisely when valuation is hardest, namely first-time sales, and thus offers new insights for both academic research and practice in art market valuation.
Authors: Yufan He, Pengfei Guo, Mengya Xu, Zhaoshuo Li, Andriy Myronenko, Dillan Imans, Bingjie Liu, Dongren Yang, Mingxue Gu, Yongnan Ji, Yueming Jin, Ren Zhao, Baiyong Shen, Daguang Xu
Abstract: Data scarcity remains a fundamental barrier to achieving fully autonomous surgical robots. While large scale vision language action (VLA) models have shown impressive generalization in household and industrial manipulation by leveraging paired video action data from diverse domains, surgical robotics suffers from the paucity of datasets that include both visual observations and accurate robot kinematics. In contrast, vast corpora of surgical videos exist, but they lack corresponding action labels, preventing direct application of imitation learning or VLA training. In this work, we aim to alleviate this problem by learning policy models from SurgWorld, a world model designed for surgical physical AI. We curated the Surgical Action Text Alignment (SATA) dataset with detailed action description specifically for surgical robots. Then we built SurgeWorld based on the most advanced physical AI world model and SATA. It's able to generate diverse, generalizable and realistic surgery videos. We are also the first to use an inverse dynamics model to infer pseudokinematics from synthetic surgical videos, producing synthetic paired video action data. We demonstrate that a surgical VLA policy trained with these augmented data significantly outperforms models trained only on real demonstrations on a real surgical robot platform. Our approach offers a scalable path toward autonomous surgical skill acquisition by leveraging the abundance of unlabeled surgical video and generative world modeling, thus opening the door to generalizable and data efficient surgical robot policies.
Authors: Will Sebelik-Lassiter, Evan Schubert, Muhammad Alliyu, Quentin Robbins, Excel Olatunji, Mustafa Barry
Abstract: Intro: Vocal cord ultrasound (VCUS) has emerged as a less invasive and better tolerated examination technique, but its accuracy is operator dependent. This research aims to apply a machine learning-assisted algorithm to automatically identify the vocal cords and distinguish normal vocal cord images from vocal cord paralysis (VCP). Methods: VCUS videos were acquired from 30 volunteers, which were split into still frames and cropped to a uniform size. Healthy and simulated VCP images were used as training data for vocal cord segmentation and VCP classification models. Results: The vocal cord segmentation model achieved a validation accuracy of 96%, while the best classification model (VIPRnet) achieved a validation accuracy of 99%. Conclusion: Machine learning-assisted analysis of VCUS shows great promise in improving diagnostic accuracy over operator-dependent human interpretation.
Authors: Qiang Sun, Zongcheng Ji, Yinlong Xiao, Peng Chang, Jun Yu
Abstract: Generating medical reports from chest X-ray images is a critical and time-consuming task for radiologists, especially in emergencies. To alleviate the stress on radiologists and reduce the risk of misdiagnosis, numerous research efforts have been dedicated to automatic medical report generation in recent years. Most recent studies have developed methods that represent images by utilizing various medical metadata, such as the clinical document history of the current patient and the medical graphs constructed from retrieved reports of other similar patients. However, all existing methods integrate additional metadata representations with visual representations through a simple "Add and LayerNorm" operation, which suffers from the information asymmetry problem due to the distinct distributions between them. In addition, chest X-ray images are usually represented using pre-trained models based on natural domain images, which exhibit an obvious domain gap between general and medical domain images. To this end, we propose a novel approach called Enhanced Image Representations (EIR) for generating accurate chest X-ray reports. We utilize cross-modal transformers to fuse metadata representations with image representations, thereby effectively addressing the information asymmetry problem between them, and we leverage medical domain pre-trained models to encode medical images, effectively bridging the domain gap for image representation. Experimental results on the widely used MIMIC and Open-I datasets demonstrate the effectiveness of our proposed method.
Authors: Sheng-Kai Chen, Jie-Yu Chao, Jr-Yu Chang, Po-Lien Wu, Po-Chiang Lin
Abstract: Visual Simultaneous Localization and Mapping (vSLAM) systems encounter substantial challenges in dynamic environments where moving objects compromise tracking accuracy and map consistency. This paper introduces PCR-ORB (Point Cloud Refinement ORB), an enhanced ORB-SLAM3 framework that integrates deep learning-based point cloud refinement to mitigate dynamic object interference. Our approach employs YOLOv8 for semantic segmentation combined with CUDA-accelerated processing to achieve real-time performance. The system implements a multi-stage filtering strategy encompassing ground plane estimation, sky region removal, edge filtering, and temporal consistency validation. Comprehensive evaluation on the KITTI dataset (sequences 00-09) demonstrates performance characteristics across different environmental conditions and scene types. Notable improvements are observed in specific sequences, with sequence 04 achieving 25.9% improvement in ATE RMSE and 30.4% improvement in ATE median. However, results show mixed performance across sequences, indicating scenario-dependent effectiveness. The implementation provides insights into dynamic object filtering challenges and opportunities for robust navigation in complex environments.
Authors: Huan-ang Gao, Zikang Zhang, Tianwei Luo, Kaisen Yang, Xinzhe Juan, Jiahao Qiu, Tianxing Chen, Bingxiang He, Hao Zhao, Hao Zhou, Shilong Liu, Mengdi Wang
Abstract: Large Language Model (LLM) agents, while proficient in the digital realm, face a significant gap in physical-world deployment due to the challenge of forming and maintaining a robust spatial mental model. We identify three core cognitive challenges hindering this transition: spatial reasoning, long-horizon state tracking via mental simulation, and active exploration under partial observation. To isolate and evaluate these faculties, we introduce CubeBench, a novel generative benchmark centered on the Rubik's Cube. CubeBench uses a three-tiered diagnostic framework that progressively assesses agent capabilities, from foundational state tracking with full symbolic information to active exploration with only partial visual data. Our experiments on leading LLMs reveal critical limitations, including a uniform 0.00% pass rate on all long-horizon tasks, exposing a fundamental failure in long-term planning. We also propose a diagnostic framework to isolate these cognitive bottlenecks by providing external solver tools. By analyzing the failure modes, we provide key insights to guide the development of more physically-grounded intelligent agents.
Authors: Jiafeng Liang, Hao Li, Chang Li, Jiaqi Zhou, Shixin Jiang, Zekun Wang, Changkai Ji, Zhihao Zhu, Runxuan Liu, Tao Ren, Jinlan Fu, See-Kiong Ng, Xia Liang, Ming Liu, Bing Qin
Abstract: Memory serves as the pivotal nexus bridging past and future, providing both humans and AI systems with invaluable concepts and experience to navigate complex tasks. Recent research on autonomous agents has increasingly focused on designing efficient memory workflows by drawing on cognitive neuroscience. However, constrained by interdisciplinary barriers, existing works struggle to assimilate the essence of human memory mechanisms. To bridge this gap, we systematically synthesizes interdisciplinary knowledge of memory, connecting insights from cognitive neuroscience with LLM-driven agents. Specifically, we first elucidate the definition and function of memory along a progressive trajectory from cognitive neuroscience through LLMs to agents. We then provide a comparative analysis of memory taxonomy, storage mechanisms, and the complete management lifecycle from both biological and artificial perspectives. Subsequently, we review the mainstream benchmarks for evaluating agent memory. Additionally, we explore memory security from dual perspectives of attack and defense. Finally, we envision future research directions, with a focus on multimodal memory systems and skill acquisition.
Authors: Mohammad Nasirzadeh, Jafar Tahmoresnezhad, Parviz Rashidi-Khazaee
Abstract: Log anomaly detection is crucial for preserving the security of operating systems. Depending on the source of log data collection, various information is recorded in logs that can be considered log modalities. In light of this intuition, unimodal methods often struggle by ignoring the different modalities of log data. Meanwhile, multimodal methods fail to handle the interactions between these modalities. Applying multimodal sentiment analysis to log anomaly detection, we propose CoLog, a framework that collaboratively encodes logs utilizing various modalities. CoLog utilizes collaborative transformers and multi-head impressed attention to learn interactions among several modalities, ensuring comprehensive anomaly detection. To handle the heterogeneity caused by these interactions, CoLog incorporates a modality adaptation layer, which adapts the representations from different log modalities. This methodology enables CoLog to learn nuanced patterns and dependencies within the data, enhancing its anomaly detection capabilities. Extensive experiments demonstrate CoLog's superiority over existing state-of-the-art methods. Furthermore, in detecting both point and collective anomalies, CoLog achieves a mean precision of 99.63%, a mean recall of 99.59%, and a mean F1 score of 99.61% across seven benchmark datasets for log-based anomaly detection. The comprehensive detection capabilities of CoLog make it highly suitable for cybersecurity, system monitoring, and operational efficiency. CoLog represents a significant advancement in log anomaly detection, providing a sophisticated and effective solution to point and collective anomaly detection through a unified framework and a solution to the complex challenges automatic log data analysis poses. We also provide the implementation of CoLog at https://github.com/NasirzadehMoh/CoLog.
Authors: Taha Emre, Arunava Chakravarty, Thomas Pinetz, Dmitrii Lachinov, Martin J. Menten, Hendrik Scholl, Sobha Sivaprasad, Daniel Rueckert, Andrew Lotery, Stefan Sacu, Ursula Schmidt-Erfurth, Hrvoje Bogunovi\'c
Abstract: Temporally aware image representations are crucial for capturing disease progression in 3D volumes of longitudinal medical datasets. However, recent state-of-the-art self-supervised learning approaches like Masked Autoencoding (MAE), despite their strong representation learning capabilities, lack temporal awareness. In this paper, we propose STAMP (Stochastic Temporal Autoencoder with Masked Pretraining), a Siamese MAE framework that encodes temporal information through a stochastic process by conditioning on the time difference between the 2 input volumes. Unlike deterministic Siamese approaches, which compare scans from different time points but fail to account for the inherent uncertainty in disease evolution, STAMP learns temporal dynamics stochastically by reframing the MAE reconstruction loss as a conditional variational inference objective. We evaluated STAMP on two OCT and one MRI datasets with multiple visits per patient. STAMP pretrained ViT models outperformed both existing temporal MAE methods and foundation models on different late stage Age-Related Macular Degeneration and Alzheimer's Disease progression prediction which require models to learn the underlying non-deterministic temporal dynamics of the diseases.
Authors: Daiki Shiono, Shumpei Miyawaki, Ryota Tanaka, Jun Suzuki
Abstract: Following the initial flourishing of large language models (LLMs), there has been a surge in proposed large vision-language models (LVLMs) that integrate LLMs with vision capabilities. However, it has been observed that LVLMs, after tuning to visual instruction using commonly used training datasets, often fail to exhibit the instruction-following ability that was present in the LLM before integration, leading to results in which they do not follow task instructions as expected. This study quantitatively demonstrates that LVLMs' instruction-following ability declines after fine-tuning and analyzes its underlying causes. In particular, we constructed new training datasets highlighting whether the output format is specified. Then, we investigated how explicitly indicating the output format during fine-tuning affects LVLMs' instruction-following ability. Our quantitative evaluation confirmed that LVLMs' instruction-following ability declines after fine-tuning with commonly used datasets. Furthermore, we found that LVLMs trained with datasets, including instructions on output format, tend to follow instructions more accurately than models that do not. These findings suggest that including samples with instructions on output format during (visual) instruction tuning may help mitigate the decline in instruction-following abilities.
Authors: Zhe Li, Cheng Chi, Yangyang Wei, Boan Zhu, Tao Huang, Zhenguo Sun, Yibo Peng, Pengwei Wang, Zhongyuan Wang, Fangzhou Liu, Chang Xu, Shanghang Zhang
Abstract: Humans learn locomotion through visual observation, interpreting visual content first before imitating actions. However, state-of-the-art humanoid locomotion systems rely on either curated motion capture trajectories or sparse text commands, leaving a critical gap between visual understanding and control. Text-to-motion methods suffer from semantic sparsity and staged pipeline errors, while video-based approaches only perform mechanical pose mimicry without genuine visual understanding. We propose RoboMirror, the first retargeting-free video-to-locomotion framework embodying "understand before you imitate". Leveraging VLMs, it distills raw egocentric/third-person videos into visual motion intents, which directly condition a diffusion-based policy to generate physically plausible, semantically aligned locomotion without explicit pose reconstruction or retargeting. Extensive experiments validate the effectiveness of RoboMirror, it enables telepresence via egocentric videos, drastically reduces third-person control latency by 80%, and achieves a 3.7% higher task success rate than baselines. By reframing humanoid control around video understanding, we bridge the visual understanding and action gap.
Authors: Jichen Feng, Yifan Zhang, Chenggong Zhang, Yifu Lu, Shilong Liu, Mengdi Wang
Abstract: Language agents increasingly require persistent worlds in which they can act, remember, and learn. Existing approaches sit at two extremes: conventional web frameworks provide reliable but fixed contexts backed by databases, while fully generative world models aim for unlimited environments at the expense of controllability and practical engineering. In this work, we introduce the Web World Model (WWM), a middle ground where world state and ``physics'' are implemented in ordinary web code to ensure logical consistency, while large language models generate context, narratives, and high-level decisions on top of this structured latent state. We build a suite of WWMs on a realistic web stack, including an infinite travel atlas grounded in real geography, fictional galaxy explorers, web-scale encyclopedic and narrative worlds, and simulation- and game-like environments. Across these systems, we identify practical design principles for WWMs: separating code-defined rules from model-driven imagination, representing latent state as typed web interfaces, and utilizing deterministic generation to achieve unlimited but structured exploration. Our results suggest that web stacks themselves can serve as a scalable substrate for world models, enabling controllable yet open-ended environments. Project Page: https://github.com/Princeton-AI2-Lab/Web-World-Models.
URLs: https://github.com/Princeton-AI2-Lab/Web-World-Models.
Authors: Kai Cheng, Xinzhe Li, Lijuan Che
Abstract: The development of unsupervised Video Anomaly Detection (VAD) relies on technologies in the field of signal processing. Since the anomaly is quite ambiguous and unbounded, different detection demands may often be raised even in one scenario. Thus, we propose to design the configurable VAD with flexible solutions targeting to solve the issue that previous methods have to train their models from scratch and waste resources when detection demands even change slightly. Moreover, we also design a dataset with good compatibility to evaluate the VAD performance when changes happen in detection demands. Besides, videos contain important information regarding continuous changes in the object's appearance and motion. Thus, we also propose a module to establish the multi-scale spatial-temporal coherence, which improves the accuracy and has the ability to dynamically adjust and accurately capture spatial-temporal normal patterns. Experiments show that our method not only models coherence effectively but also has better configurable ability.
Authors: Milad Abdollahzadeh, Guimeng Liu, Touba Malekzadeh, Christopher T. H. Teo, Keshigeyan Chandrasegaran, Ngai-Man Cheung
Abstract: Generative modeling in machine learning aims to synthesize new data samples that are statistically similar to those observed during training. While conventional generative models such as GANs and diffusion models typically assume access to large and diverse datasets, many real-world applications (e.g. in medicine, satellite imaging, and artistic domains) operate under limited data availability and strict constraints. In this survey, we examine Generative Modeling under Data Constraint (GM-DC), which includes limited-data, few-shot, and zero-shot settings. We present a unified perspective on the key challenges in GM-DC, including overfitting, frequency bias, and incompatible knowledge transfer, and discuss how these issues impact model performance. To systematically analyze this growing field, we introduce two novel taxonomies: one categorizing GM-DC tasks (e.g. unconditional vs. conditional generation, cross-domain adaptation, and subject-driven modeling), and another organizing methodological approaches (e.g. transfer learning, data augmentation, meta-learning, and frequency-aware modeling). Our study reviews over 230 papers, offering a comprehensive view across generative model types and constraint scenarios. We further analyze task-approach-method interactions using a Sankey diagram and highlight promising directions for future work, including adaptation of foundation models, holistic evaluation frameworks, and data-centric strategies for sample selection. This survey provides a timely and practical roadmap for researchers and practitioners aiming to advance generative modeling under limited data. Project website: https://sutd-visual-computing-group.github.io/gmdc-survey/.
URLs: https://sutd-visual-computing-group.github.io/gmdc-survey/.
Authors: Nico Baumgart, Markus Lange-Hegermann, Mike M\"ucke
Abstract: In industrial manufacturing, deploying deep learning models for visual inspection is mostly hindered by the high and often intractable cost of collecting and annotating large-scale training datasets. While image synthesis from 3D CAD models is a common solution, the individual techniques of domain and rendering randomization to create rich synthetic training datasets have been well studied mainly in simple domains. Hence, their effectiveness on complex industrial tasks with densely arranged and similar objects remains unclear. In this paper, we investigate the sim-to-real generalization performance of standard object detectors on the complex industrial application of terminal strip object detection, carefully combining randomization and domain knowledge. We describe step-by-step the creation of our image synthesis pipeline that achieves high realism with minimal implementation effort and explain how this approach could be transferred to other industrial settings. Moreover, we created a dataset comprising 30.000 synthetic images and 300 manually annotated real images of terminal strips, which is publicly available for reference and future research. To provide a baseline as a lower bound of the expectable performance in these challenging industrial parts detection tasks, we show the sim-to-real generalization performance of standard object detectors on our dataset based on a fully synthetic training. While all considered models behave similarly, the transformer-based DINO model achieves the best score with 98.40 % mean average precision on the real test set, demonstrating that our pipeline enables high quality detections in complex industrial environments from existing CAD data and with a manageable image synthesis effort.
Authors: Vlas Zyrianov, Henry Che, Zhijian Liu, Shenlong Wang
Abstract: We present LidarDM, a novel LiDAR generative model capable of producing realistic, layout-aware, physically plausible, and temporally coherent LiDAR videos. LidarDM stands out with two unprecedented capabilities in LiDAR generative modeling: (i) LiDAR generation guided by driving scenarios, offering significant potential for autonomous driving simulations, and (ii) 4D LiDAR point cloud generation, enabling the creation of realistic and temporally coherent sequences. At the heart of our model is a novel integrated 4D world generation framework. Specifically, we employ latent diffusion models to generate the 3D scene, combine it with dynamic actors to form the underlying 4D world, and subsequently produce realistic sensory observations within this virtual environment. Our experiments indicate that our approach outperforms competing algorithms in realism, temporal coherency, and layout consistency. We additionally show that LidarDM can be used as a generative world model simulator for training and testing perception models.
Authors: Tiange Luo, Justin Johnson, Honglak Lee
Abstract: Scalable annotation approaches are crucial for constructing extensive 3D-text datasets, facilitating a broader range of applications. However, existing methods sometimes lead to the generation of hallucinated captions, compromising caption quality. This paper explores the issue of hallucination in 3D object captioning, with a focus on Cap3D method, which renders 3D objects into 2D views for captioning using pre-trained models. We pinpoint a major challenge: certain rendered views of 3D objects are atypical, deviating from the training data of standard image captioning models and causing hallucinations. To tackle this, we present DiffuRank, a method that leverages a pre-trained text-to-3D model to assess the alignment between 3D objects and their 2D rendered views, where the view with high alignment closely represent the object's characteristics. By ranking all rendered views and feeding the top-ranked ones into GPT4-Vision, we enhance the accuracy and detail of captions, enabling the correction of 200k captions in the Cap3D dataset and extending it to 1 million captions across Objaverse and Objaverse-XL datasets. Additionally, we showcase the adaptability of DiffuRank by applying it to pre-trained text-to-image models for a Visual Question Answering task, where it outperforms the CLIP model.
Authors: Jiaqi Yang, Nitish Mehta, Xiaoling Hu, Chao Chen, Chia-Ling Tsai
Abstract: Accurate segmentation of Optical Coherence Tomography (OCT) images is crucial for diagnosing and monitoring retinal diseases. However, the labor-intensive nature of pixel-level annotation limits the scalability of supervised learning for large datasets. Weakly Supervised Semantic Segmentation (WSSS) offers a promising alternative by using weaker forms of supervision, such as image-level labels, to reduce the annotation burden. Despite its advantages, weak supervision inherently carries limited information. We propose a novel WSSS framework with only image-level labels for OCT lesion segmentation that integrates structural and text-driven guidance to produce high-quality, pixel-level pseudo labels. The framework employs two visual processing modules: one that processes the original OCT images and another that operates on layer segmentations augmented with anomalous signals, enabling the model to associate lesions with their corresponding anatomical layers. Complementing these visual cues, we leverage large-scale pretrained models to provide two forms of textual guidance: label-derived descriptions that encode local semantics, and domain-agnostic synthetic descriptions that, although expressed in natural image terms, capture spatial and relational semantics useful for generating globally consistent representations. By fusing these visual and textual features in a multi-modal framework, our method aligns semantic meaning with structural relevance, thereby improving lesion localization and segmentation performance. Experiments on three OCT datasets demonstrate state-of-the-art results, highlighting its potential to advance diagnostic accuracy and efficiency in medical imaging.
Authors: Zhihao Sun, Haoran Jiang, Haoran Chen, Yixin Cao, Xipeng Qiu, Zuxuan Wu, Yu-Gang Jiang
Abstract: Multimodal large language models have unlocked new possibilities for various multimodal tasks. However, their potential in image manipulation detection remains unexplored. When directly applied to the IMD task, M-LLMs often produce reasoning texts that suffer from hallucinations and overthinking. To address this, we propose ForgerySleuth, which leverages M-LLMs to perform comprehensive clue fusion and generate segmentation outputs indicating specific regions that are tampered with. Moreover, we construct the ForgeryAnalysis dataset through the Chain-of-Clues prompt, which includes analysis and reasoning text to upgrade the image manipulation detection task. A data engine is also introduced to build a larger-scale dataset for the pre-training phase. Our extensive experiments demonstrate the effectiveness of ForgeryAnalysis and show that ForgerySleuth significantly outperforms existing methods in generalization, robustness, and explainability.
Authors: Pritesh Prakash, Anoop Kumar Rai
Abstract: Aging presents a significant challenge in face recognition, as changes in skin texture and tone can alter facial features over time, making it particularly difficult to compare images of the same individual taken years apart, such as in long-term identification scenarios. Transformer networks have the strength to preserve sequential spatial relationships caused by aging effect. This paper presents a technique for loss evaluation that uses a transformer network as an additive loss in the face recognition domain. The standard metric loss function typically takes the final embedding of the main CNN backbone as its input. Here, we employ a transformer-metric loss, a combined approach that integrates both transformer-loss and metric-loss. This research intends to analyze the transformer behavior on the convolution output when the CNN outcome is arranged in a sequential vector. These sequential vectors have the potential to overcome the texture or regional structure referred to as wrinkles or sagging skin affected by aging. The transformer encoder takes input from the contextual vectors obtained from the final convolution layer of the network. The learned features can be more age-invariant, complementing the discriminative power of the standard metric loss embedding. With this technique, we use transformer loss with various base metric-loss functions to evaluate the effect of the combined loss functions. We observe that such a configuration allows the network to achieve SoTA results in LFW and age-variant datasets (CA-LFW and AgeDB). This research expands the role of transformers in the machine vision domain and opens new possibilities for exploring transformers as a loss function.
Authors: Yixiong Fang, Ziran Yang, Zhaorun Chen, Zhuokai Zhao, Jiawei Zhou
Abstract: Large vision-language models (LVLMs) excel at multimodal tasks but are prone to misinterpreting visual inputs, often resulting in hallucinations and unreliable outputs. We present DROPOUT DECODING, a novel inference-time approach that quantifies the uncertainty of visual tokens and selectively masks uncertain tokens to improve decoding. Our method measures the uncertainty of each visual token by projecting it onto the text space and decomposing it into aleatoric and epistemic components. Specifically, we focus on epistemic uncertainty, which captures perception-related errors more effectively. Inspired by dropout regularization, we introduce uncertainty-guided token dropout, which applies the dropout principle to input visual tokens instead of model parameters, and during inference rather than training. By aggregating predictions from an ensemble of masked decoding contexts, we can robustly mitigate errors arising from visual token misinterpretations. Evaluations on benchmarks including CHAIR, THRONE, and MMBench demonstrate that DROPOUT DECODING significantly reduces object hallucinations (OH) and enhances both reliability and quality of LVLM outputs across diverse visual contexts. Code is released at https://github.com/kigb/DropoutDecoding.
Authors: Kavana Venkatesh, Yusuf Dalva, Ismini Lourentzou, Pinar Yanardag
Abstract: Despite impressive visual fidelity, current text-to-image (T2I) diffusion models struggle to depict rare, complex, or culturally nuanced concepts due to training data limitations. We introduce RAVEL, a training-free framework that significantly improves rare concept generation, context-driven image editing, and self-correction by integrating graph-based retrieval-augmented generation (RAG) into diffusion pipelines. Unlike prior RAG and LLM-enhanced methods reliant on visual exemplars, static captions or pre-trained knowledge of models, RAVEL leverages structured knowledge graphs to retrieve compositional, symbolic, and relational context, enabling nuanced grounding even in the absence of visual priors. To further refine generation quality, we propose SRD, a novel self-correction module that iteratively updates prompts via multi-aspect alignment feedback, enhancing attribute accuracy, narrative coherence, and semantic fidelity. Our framework is model-agnostic and compatible with leading diffusion models including Stable Diffusion XL, Flux, and DALL-E 3. We conduct extensive evaluations across three newly proposed benchmarks - MythoBench, Rare-Concept-1K, and NovelBench. RAVEL also consistently outperforms SOTA methods across perceptual, alignment, and LLM-as-a-Judge metrics. These results position RAVEL as a robust paradigm for controllable and interpretable T2I generation in long-tail domains.
Authors: Bi'an Du, Wei Hu, Renjie Liao
Abstract: Consistency Models (CMs) have significantly accelerated the sampling process in diffusion models, yielding impressive results in synthesizing high-resolution images. To explore and extend these advancements to point-cloud-based 3D shape generation, we propose a novel Multi-scale Latent Point Consistency Model (MLPCM). Our MLPCM follows a latent diffusion framework and introduces hierarchical levels of latent representations, ranging from point-level to super-point levels, each corresponding to a different spatial resolution. We design a multi-scale latent integration module along with 3D spatial attention to effectively denoise the point-level latent representations conditioned on those from multiple super-point levels. Additionally, we propose a latent consistency model, learned through consistency distillation, that compresses the prior into a one-step generator. This significantly improves sampling efficiency while preserving the performance of the original teacher model. Extensive experiments on standard benchmarks ShapeNet and ShapeNet-Vol demonstrate that MLPCM achieves a 100x speedup in the generation process, while surpassing state-of-the-art diffusion models in terms of both shape quality and diversity.
Authors: Xindi Wu, Mengzhou Xia, Rulin Shao, Zhiwei Deng, Pang Wei Koh, Olga Russakovsky
Abstract: Training vision-language models via instruction tuning relies on large data mixtures spanning diverse tasks and domains, yet these mixtures frequently include redundant information that increases computational costs without proportional gains. Existing methods typically rely on task-agnostic heuristics to estimate data importance, limiting their effectiveness across tasks. We introduce ICONS, a gradient-based Influence CONsensus approach for vision-language data Selection. Our method leverages first-order training dynamics to estimate each example's influence on validation performance, then aggregates these estimates across tasks via majority voting. This cross-task consensus identifies consistently valuable data points while mitigating score calibration and outlier sensitivity, enabling robust and scalable data selection for diverse multitask mixtures. Models trained on our selected 20% data subset from LLAVA-665K (respectively: from CAMBRIAN-7M, from VISION-FLAN-186K) retain 98.6% (respectively: 98.8%, 99.8%) of full-dataset performance. We demonstrate that our selected data generalizes to unseen tasks and model architectures, and release three compact subsets LLAVA-ICONS-133K, CAMBRIAN-ICONS-1.4M, and VISION-FLAN-ICONS-37K for efficient vision-language model development.
Authors: Qifeng Zhou, Wenliang Zhong, Thao M. Dang, Hehuan Ma, Saiyang Na, Yuzhi Guo, Junzhou Huang
Abstract: The integration of Artificial Intelligence (AI) into pathology faces a fundamental challenge: black-box predictive models lack transparency, while generative approaches risk clinical hallucination. A case-based retrieval paradigm offers a more interpretable alternative for clinical adoption. However, current SOTA models are constrained by dual-encoder architectures that cannot process the composed modality of real-world clinical queries. We formally define the task of Pathology Composed Retrieval (PCR). However, progress in this newly defined task is blocked by two critical challenges: (1) Multimodal Large Language Models (MLLMs) offer the necessary deep-fusion architecture but suffer from a critical Task Mismatch and Domain Mismatch. (2) No benchmark exists to evaluate such compositional queries. To solve these challenges, we propose HOMIE, a systematic framework that transforms a general MLLM into a specialized retrieval expert. HOMIE resolves the dual mismatch via a two-stage process: a retrieval-adaptation stage to solve the task mismatch, and a pathology-specific tuning stage, featuring a progressive knowledge curriculum, pathology specfic stain and native resolution processing, to solve the domain mismatch. We also introduce the PCR Benchmark, a benchmark designed to evaluate composed retrieval in pathology. Experiments show that HOMIE, trained only on public data, matches SOTA performance on traditional retrieval tasks and outperforms all baselines on the newly defined PCR task.
Authors: Jia Yu, Yan Zhu, Peiyao Fu, Tianyi Chen, Junbo Huang, Quanlin Li, Pinghong Zhou, Zhihua Wang, Fei Wu, Shuo Wang, Xian Yang
Abstract: Colorectal cancer (CRC) is a significant global health concern, and early detection through screening plays a critical role in reducing mortality. While deep learning models have shown promise in improving polyp detection, classification, and segmentation, their generalization across diverse clinical environments, particularly with out-of-distribution (OOD) data, remains a challenge. Multi-center datasets like PolypGen have been developed to address these issues, but their collection is costly and time-consuming. Traditional data augmentation techniques provide limited variability, failing to capture the complexity of medical images. Diffusion models have emerged as a promising solution for generating synthetic polyp images, but the image generation process in current models mainly relies on segmentation masks as the condition, limiting their ability to capture the full clinical context. To overcome these limitations, we propose a Progressive Spectrum Diffusion Model (PSDM) that integrates diverse clinical annotations-such as segmentation masks, bounding boxes, and colonoscopy reports-by transforming them into compositional prompts. These prompts are organized into coarse and fine components, allowing the model to capture both broad spatial structures and fine details, generating clinically accurate synthetic images. By augmenting training data with PSDM-generated samples, our model significantly improves polyp detection, classification, and segmentation. For instance, on the PolypGen dataset, PSDM increases the F1 score by 2.12% and the mean average precision by 3.09%, demonstrating superior performance in OOD scenarios and enhanced generalization.
Authors: Yimin Zhu, Lincoln Linlin Xu
Abstract: Data augmentation effectively addresses the imbalanced-small sample data (ISSD) problem in hyperspectral image classification (HSIC). While most methodologies extend features in the latent space, few leverage text-driven generation to create realistic and diverse samples. Recently, text-guided diffusion models have gained significant attention due to their ability to generate highly diverse and high-quality images based on text prompts in natural image synthesis. Motivated by this, this paper proposes Txt2HSI-LDM(VAE), a novel language-informed hyperspectral image synthesis method to address the ISSD in HSIC. The proposed approach uses a denoising diffusion model, which iteratively removes Gaussian noise to generate hyperspectral samples conditioned on textual descriptions. First, to address the high-dimensionality of hyperspectral data, a universal variational autoencoder (VAE) is designed to map the data into a low-dimensional latent space, which provides stable features and reduces the inference complexity of diffusion model. Second, a semi-supervised diffusion model is designed to fully take advantage of unlabeled data. Random polygon spatial clipping (RPSC) and uncertainty estimation of latent feature (LF-UE) are used to simulate the varying degrees of mixing. Third, the VAE decodes HSI from latent space generated by the diffusion model with the language conditions as input. In our experiments, we fully evaluate synthetic samples' effectiveness from statistical characteristics and data distribution in 2D-PCA space. Additionally, visual-linguistic cross-attention is visualized on the pixel level to prove that our proposed model can capture the spatial layout and geometry of the generated data. Experiments demonstrate that the performance of the proposed Txt2HSI-LDM(VAE) surpasses the classical backbone models, state-of-the-art CNNs, and semi-supervised methods.
Authors: Zhehui Wu, Yong Chen, Naoto Yokoya, Wei He
Abstract: Hyperspectral images (HSIs) often suffer from diverse and unknown degradations during imaging, leading to severe spectral and spatial distortions. Existing HSI restoration methods typically rely on specific degradation assumptions, limiting their effectiveness in complex scenarios. In this paper, we propose \textbf{MP-HSIR}, a novel multi-prompt framework that effectively integrates spectral, textual, and visual prompts to achieve universal HSI restoration across diverse degradation types and intensities. Specifically, we develop a prompt-guided spatial-spectral transformer, which incorporates spatial self-attention and a prompt-guided dual-branch spectral self-attention. Since degradations affect spectral features differently, we introduce spectral prompts in the local spectral branch to provide universal low-rank spectral patterns as prior knowledge for enhancing spectral reconstruction. Furthermore, the text-visual synergistic prompt fuses high-level semantic representations with fine-grained visual features to encode degradation information, thereby guiding the restoration process. Extensive experiments on 9 HSI restoration tasks, including all-in-one scenarios, generalization tests, and real-world cases, demonstrate that MP-HSIR not only consistently outperforms existing all-in-one methods but also surpasses state-of-the-art task-specific approaches across multiple tasks. The code and models are available at https://github.com/ZhehuiWu/MP-HSIR.
Authors: Jing Wang, Fengzhuo Zhang, Xiaoli Li, Vincent Y. F. Tan, Tianyu Pang, Chao Du, Aixin Sun, Zhuoran Yang
Abstract: Auto-Regressive Video Diffusion Models (AR-VDMs) have shown strong capabilities in generating long, photorealistic videos, but suffer from two key limitations: (i) history forgetting, where the model loses track of previously generated content, and (ii) temporal degradation, where frame quality deteriorates over time. Yet a rigorous theoretical analysis of these phenomena is lacking, and existing empirical understanding remains insufficiently grounded. In this paper, we introduce Meta-ARVDM, a unified analytical framework that studies both errors through the shared autoregressive structure of AR-VDMs. We show that history forgetting is characterized by the conditional mutual information between the generated output and preceding frames, conditioned on inputs, and prove that incorporating more past frames monotonically alleviates history forgetting, thereby theoretically justifying a common belief in existing works. Moreover, our theory reveals that standard metrics fail to capture this effect, motivating a new evaluation protocol based on a ``needle-in-a-haystack'' task in closed-ended environments (DMLab and Minecraft). We further show that temporal degradation can be quantified by the cumulative sum of per-step errors, enabling prediction of degradation for different schedulers without video rollout. Finally, our evaluation uncovers a strong empirical correlation between history forgetting and temporal degradation, a connection not previously reported.
Authors: Gang Wu, Junjun Jiang, Kui Jiang, Xianming Liu, Liqiang Nie
Abstract: Image restoration has witnessed significant advancements with the development of deep learning models. Transformer-based models, particularly those using window-based self-attention, have become a dominant force. However, their performance is constrained by the rigid, non-overlapping window partitioning scheme, which leads to \textit{insufficient feature interaction across windows and limited receptive fields}. This highlights the need for more adaptive and flexible attention mechanisms. In this paper, we propose the Deformable Sliding Window Transformer for Image Restoration (DSwinIR), a new attention mechanism: the {Deformable Sliding Window (DSwin) Attention}. {This mechanism introduces a token-centric and content-aware paradigm that moves beyond the grid and fixed window partition.} It comprises two complementary components. First, it replaces the rigid partitioning with a \textit{token-centric sliding window} paradigm, {making it effective at eliminating boundary artifacts}. Second, it incorporates a \textit{content-aware deformable sampling} strategy, which allows the attention mechanism to learn data-dependent offsets and actively shape its receptive field to focus on the most informative image regions. Extensive experiments show that DSwinIR achieves strong results, including state-of-the-art performance on several evaluated benchmarks. For instance, in all-in-one image restoration, our DSwinIR surpasses the most recent backbone GridFormer by 0.53 dB on the three-task benchmark and 0.87 dB on the five-task benchmark.
Authors: Yizhuo Wu, Francesco Fioranelli, Chang Gao
Abstract: Radar-based Human Activity Recognition (HAR) is an attractive alternative to wearables and cameras because it preserves privacy, and is contactless and robust to occlusions. However, dominant Convolutional Neural Network (CNN)- and Recurrent Neural Network (RNN)-based solutions are computationally intensive at deployment, and recent lightweight Vision Transformer (ViT) and State Space Model (SSM) variants still exhibit substantial complexity. In this paper, we present RadMamba, a parameter-efficient, micro-Doppler-oriented Mamba SSM tailored to radar HAR under on-sensor compute, latency, and energy constraints typical of distributed radar systems. RadMamba combines (i) channel fusion with downsampling, (ii) Doppler-aligned segmentation that preserves the physical continuity of Doppler over time, and (iii) convolutional token projections that better capture Doppler-span variations, thereby retaining temporal-Doppler structure while reducing the number of Floating-point Operations per Inference (#FLOP/Inf.). Evaluated across three datasets with different radars and types of activities, RadMamba matches the prior best 99.8% accuracy of a recent SSM-based model on the Continuous Wave (CW) radar dataset, while requiring only 1/400 of its parameters. On a dataset of non-continuous activities with Frequency Modulated Continuous Wave (FMCW) radar, RadMamba remains competitive with leading 92.0% results using about 1/10 of the parameters, and on a continuous FMCW radar dataset it surpasses methods with far more parameters by at least 3%, using only 6.7k parameters. Code: https://github.com/lab-emi/AIRHAR.
Authors: Qi Fan, Kaiqi Liu, Nian Liu, Hisham Cholakkal, Rao Muhammad Anwer, Wenbin Li, Yang Gao
Abstract: Cross-domain few-shot segmentation (CD-FSS) aims to segment objects of novel classes in new domains, which is often challenging due to the diverse characteristics of target domains and the limited availability of support data. Most CD-FSS methods redesign and retrain in-domain FSS models using abundant base data from the source domain, which are effective but costly to train. To address these issues, we propose adapting informative model structures of the well-trained FSS model for target domains by learning domain characteristics from few-shot labeled support samples during inference, thereby eliminating the need for source domain retraining. Specifically, we first adaptively identify domain-specific model structures by measuring parameter importance using a novel structure Fisher score in a data-dependent manner. Then, we progressively train the selected informative model structures with hierarchically constructed training samples, progressing from fewer to more support shots. The resulting Informative Structure Adaptation (ISA) method effectively addresses domain shifts and equips existing well-trained in-domain FSS models with flexible adaptation capabilities for new domains, eliminating the need to redesign or retrain CD-FSS models on base data. Extensive experiments validate the effectiveness of our method, demonstrating superior performance across multiple CD-FSS benchmarks. Codes are at https://github.com/fanq15/ISA.
Authors: Qinfeng Zhu, Yunxi Jiang, Lei Fan
Abstract: We propose a result-level category-specific fusion architecture called ClassWise-CRF. This architecture employs a two-stage process: first, it selects expert networks that perform well in specific categories from a pool of candidate networks using a greedy algorithm; second, it integrates the segmentation predictions of these selected networks by adaptively weighting their contributions based on their segmentation performance in each category. Inspired by Conditional Random Field (CRF), the ClassWise-CRF architecture treats the segmentation predictions from multiple networks as confidence vector fields. It leverages segmentation metrics (such as Intersection over Union) from the validation set as priors and employs an exponential weighting strategy to fuse the category-specific confidence scores predicted by each network. This fusion method dynamically adjusts the weights of each network for different categories, achieving category-specific optimization. Building on this, the architecture further optimizes the fused results using unary and pairwise potentials in CRF to ensure spatial consistency and boundary accuracy. To validate the effectiveness of ClassWise-CRF, we conducted experiments on two remote sensing datasets, LoveDA and Vaihingen, using eight classic and advanced semantic segmentation networks. The results show that the ClassWise-CRF architecture significantly improves segmentation performance: on the LoveDA dataset, the mean Intersection over Union (mIoU) metric increased by 1.00% on the validation set and by 0.68% on the test set; on the Vaihingen dataset, the mIoU improved by 0.87% on the validation set and by 0.91% on the test set. These results fully demonstrate the effectiveness and generality of the ClassWise-CRF architecture in semantic segmentation of remote sensing images. The full code is available at https://github.com/zhuqinfeng1999/ClassWise-CRF.
Authors: Hao Yang, Yan Yang, Ruikun Zhang, Liyuan Pan
Abstract: Recent advances in OpenAI's GPT-series multimodal generation models have shown remarkable capabilities in producing visually compelling images. In this work, we investigate its potential impact on the image restoration community. We provide, to the best of our knowledge, the first systematic benchmark across diverse restoration scenarios. Our evaluation shows that, while the restoration results generated by GPT-Image models are often perceptually pleasant, they tend to lack pixel-level structural fidelity compared with ground-truth references. Typical deviations include changes in image geometry, object positions or counts, and even modifications in perspective. Beyond empirical observations, we further demonstrate that outputs from GPT-Image models can act as strong visual priors, offering notable performance improvements for existing restoration networks. Using dehazing, deraining, and low-light enhancement as representative case studies, we show that integrating GPT-generated priors significantly boosts restoration quality. This study not only provides practical insights and a baseline framework for incorporating GPT-based generative priors into restoration pipelines, but also highlights new opportunities for bridging image generation models and restoration tasks. To support future research, we will release GPT-restored results.
Authors: Xuecheng Wu, Jiaxing Liu, Danlei Huang, Yifan Wang, Yunyun Shi, Kedi Chen, Junxiao Xue, Yang Liu, Chunlin Chen, Hairong Dong, Dingkang Yang
Abstract: Visual-Interleaved Chain-of-Thought (VI-CoT) enables Multi-modal Large Language Models (MLLMs) to continually update their understanding and decision space based on step-wise intermediate visual states (IVS), much like a human would, which has demonstrated impressive success in various tasks, thereby leading to emerged advancements in related downstream benchmarks. Despite promising progress, current benchmarks provide models with relatively fixed IVS, rather than free-style IVS, whch might forcibly distort the original thinking trajectories, failing to evaluate their intrinsic reasoning capabilities. More importantly, existing benchmarks neglect to systematically explore the impact factors that IVS would impart to the untamed reasoning performance. To tackle above gaps, we introduce a specialized benchmark termed ViC-Bench, consisting of four representive tasks, i.e., maze navigation, jigsaw puzzle, embodied long-horizon planning, as well as complex counting, where each task has dedicated free-style IVS generation pipeline supporting adaptive function calls. To systematically examine VI-CoT capability, we propose a thorough evaluation suite incorporating a progressive three-stage strategy with targeted new metrics. Besides, we establish Incremental Prompting Information Injection strategy to ablatively explore the prompting factors for VI-CoT. We extensively conduct evaluations for 18 advanced MLLMs, revealing key insights into their VI-CoT capability. The introduced ViC-Bench has been made publicly available at Huggingface.
Authors: Cainan Davidson, Deva Ramanan, Neehar Peri
Abstract: Autonomous Vehicles (AVs) collect and pseudo-label terabytes of multi-modal data localized to HD maps during normal fleet testing. However, identifying interesting and safety-critical scenarios from uncurated driving logs remains a significant challenge. Traditional scenario mining techniques are error-prone and prohibitively time-consuming, often relying on hand-crafted structured queries. In this work, we revisit spatio-temporal scenario mining through the lens of recent vision-language models (VLMs) to detect whether a described scenario occurs in a driving log and, if so, precisely localize it in both time and space. To address this problem, we introduce RefAV, a large-scale dataset of 10,000 diverse natural language queries that describe complex multi-agent interactions relevant to motion planning derived from 1000 driving logs in the Argoverse 2 Sensor dataset. We evaluate several referential multi-object trackers and present an empirical analysis of our baselines. Notably, we find that naively repurposing off-the-shelf VLMs yields poor performance, suggesting that scenario mining presents unique challenges. Lastly, we discuss our recently held competition and share insights from the community. Our code and dataset are available at https://github.com/CainanD/RefAV/ and https://argoverse.github.io/user-guide/tasks/scenario_mining.html
URLs: https://github.com/CainanD/RefAV/, https://argoverse.github.io/user-guide/tasks/scenario_mining.html
Authors: Yi Liao, Ugochukwu Ejike Akpudo, Jue Zhang, Yongsheng Gao, Jun Zhou, Wenyi Zeng, Weichuan Zhang
Abstract: Visual explanation maps enhance the trustworthiness of decisions made by deep learning models and offer valuable guidance for developing new algorithms in image recognition tasks. Class activation maps (CAM) and their variants (e.g., Grad-CAM and Relevance-CAM) have been extensively employed to explore the interpretability of softmax-based convolutional neural networks, which require a fully connected layer as the classifier for decision-making. However, these methods cannot be directly applied to metric learning models, as such models lack a fully connected layer functioning as a classifier. To address this limitation, we propose a novel visual explanation method termed Similar Feature Activation Map (SFAM). This method introduces the channel-wise contribution importance score (CIS) to measure feature importance, derived from the similarity measurement between two image embeddings. The explanation map is constructed by linearly combining the proposed importance weights with the feature map from a CNN model. Quantitative and qualitative experiments show that SFAM provides highly promising interpretable visual explanations for CNN models using Euclidean distance or cosine similarity as the similarity metric.
Authors: Yake Wei, Yu Miao, Dongzhan Zhou, Di Hu
Abstract: In this paper, we reveal that most current efficient multimodal fine-tuning methods are hindered by a key limitation: they are directly borrowed from LLMs, often neglecting the intrinsic differences of multimodal scenarios and even affecting the full utilization of all modalities. Inspired by our empirical observation, we argue that unimodal adaptation and cross-modal adaptation are two essential parts for the effective fine-tuning of MLLMs. From this perspective, we propose Multimodal low-rank Adaptation (MokA), a multimodal-aware efficient fine-tuning strategy that takes multimodal characteristics into consideration. It compresses unimodal information by modality-specific parameters while explicitly enhancing cross-modal interaction, ensuring both unimodal and cross-modal adaptation. Extensive experiments cover three representative multimodal scenarios (audio-visual-text, visual-text, and speech-text), and multiple LLM backbones (LLaMA2/3, Qwen2, Qwen2.5-VL, etc). Consistent improvements indicate the efficacy and versatility of the proposed method. Ablation studies and efficiency evaluation are also conducted to fully asses our method. Overall, we think MokA provides a more targeted solution for efficient adaptation of MLLMs, paving the way for further exploration. The project page is at https://gewu-lab.github.io/MokA.
Authors: Guoyi Zhang, Guangsheng Xu, Siyang Chen, Han Wang, Xiaohu Zhang
Abstract: \textcolor{blue}{This is the pre-acceptance version, to read the final version please go to \href{https://ieeexplore.ieee.org/document/11156113}{IEEE Transactions on Geoscience and Remote Sensing on IEEE Xplore}.} Infrared small target detection (IRSTD) remains a long-standing challenge in complex backgrounds due to low signal-to-clutter ratios (SCR), diverse target morphologies, and the absence of distinctive visual cues. While recent deep learning approaches aim to learn discriminative representations, the intrinsic variability and weak priors of small targets often lead to unstable performance. In this paper, we propose a novel end-to-end IRSTD framework, termed LRRNet, which leverages the low-rank property of infrared image backgrounds. Inspired by the physical compressibility of cluttered scenes, our approach adopts a compression--reconstruction--subtraction (CRS) paradigm to directly model structure-aware low-rank background representations in the image domain, without relying on patch-based processing or explicit matrix decomposition. To the best of our knowledge, this is the first work to directly learn low-rank background structures using deep neural networks in an end-to-end manner. Extensive experiments on multiple public datasets demonstrate that LRRNet outperforms 38 state-of-the-art methods in terms of detection accuracy, robustness, and computational efficiency. Remarkably, it achieves real-time performance with an average speed of 82.34 FPS. Evaluations on the challenging NoisySIRST dataset further confirm the model's resilience to sensor noise. The source code will be made publicly available upon acceptance.
Authors: Zicheng Zhao, Kangyu Wang, Shijie Li, Rui Qian, Weiyao Lin, Huabin Liu
Abstract: Despite advancements in Video Large Language Models (Vid-LLMs) improving multimodal understanding, challenges persist in streaming video reasoning due to its reliance on contextual information. Existing paradigms feed all available historical contextual information into Vid-LLMs, resulting in a significant computational burden for visual data processing. Furthermore, the inclusion of irrelevant context distracts models from key details. This paper introduces a challenging task called Context-guided Streaming Video Reasoning (CogStream), which simulates real-world streaming video scenarios, requiring models to identify the most relevant historical contextual information to deduce answers for questions about the current stream. To support CogStream, we present a densely annotated dataset featuring extensive and hierarchical question-answer pairs, generated by a semi-automatic pipeline. Additionally, we present CogReasoner as a baseline model. It effectively tackles this task by leveraging visual stream compression and historical dialogue retrieval. Extensive experiments prove the effectiveness of this method.
Authors: Olivia Zumsteg, Nico Graf, Aaron Haeusler, Norbert Kirchgessner, Nicola Storni, Lukas Roth, Andreas Hund
Abstract: Estimating three-dimensional morphological traits such as volume from two-dimensional RGB images presents inherent challenges due to the loss of depth information, projection distortions, and occlusions under field conditions. In this work, we explore multiple approaches for non-destructive volume estimation of wheat spikes using RGB images and structured-light 3D scans as ground truth references. Wheat spike volume is promising for phenotyping as it shows high correlation with spike dry weight, a key component of fruiting efficiency. Accounting for the complex geometry of the spikes, we compare different neural network approaches for volume estimation from 2D images and benchmark them against two conventional baselines: a 2D area-based projection and a geometric reconstruction using axis-aligned cross-sections. Fine-tuned Vision Transformers (DINOv2 and DINOv3) with MLPs achieve the lowest MAPE of 5.08\% and 4.67\% and the highest correlation of 0.96 and 0.97 on six-view indoor images, outperforming fine-tuned CNNs (ResNet18 and ResNet50), wheat-specific backbones, and both baselines. When using frozen DINO backbones, deep-supervised LSTMs outperform MLPs, whereas after fine-tuning, improved high-level representations allow simple MLPs to outperform LSTMs. We demonstrate that object shape significantly impacts volume estimation accuracy, with irregular geometries such as wheat spikes posing greater challenges for geometric methods than for deep learning approaches. Fine-tuning DINOv3 on field-based single side-view images yields a MAPE of 8.39\% and a correlation of 0.90, providing a novel pipeline and a fast, accurate, and non-destructive approach for wheat spike volume phenotyping.
Authors: Chengyu Wang, Isabella Bennett, Henry Scott, Liang Zhang, Mei Chen, Hao Li, Rui Zhao
Abstract: We present MatDecompSDF, a novel framework for recovering high-fidelity 3D shapes and decomposing their physically-based material properties from multi-view images. The core challenge of inverse rendering lies in the ill-posed disentanglement of geometry, materials, and illumination from 2D observations. Our method addresses this by jointly optimizing three neural components: a neural Signed Distance Function (SDF) to represent complex geometry, a spatially-varying neural field for predicting PBR material parameters (albedo, roughness, metallic), and an MLP-based model for capturing unknown environmental lighting. The key to our approach is a physically-based differentiable rendering layer that connects these 3D properties to the input images, allowing for end-to-end optimization. We introduce a set of carefully designed physical priors and geometric regularizations, including a material smoothness loss and an Eikonal loss, to effectively constrain the problem and achieve robust decomposition. Extensive experiments on both synthetic and real-world datasets (e.g., DTU) demonstrate that MatDecompSDF surpasses state-of-the-art methods in geometric accuracy, material fidelity, and novel view synthesis. Crucially, our method produces editable and relightable assets that can be seamlessly integrated into standard graphics pipelines, validating its practical utility for digital content creation.
Authors: L'ea Dubois, Klaus Schmidt, Chengyu Wang, Ji-Hoon Park, Lin Wang, Santiago Munoz
Abstract: Current video understanding models excel at recognizing "what" is happening but fall short in high-level cognitive tasks like causal reasoning and future prediction, a limitation rooted in their lack of commonsense world knowledge. To bridge this cognitive gap, we propose a novel framework that synergistically fuses a powerful Vision Foundation Model (VFM) for deep visual perception with a Large Language Model (LLM) serving as a knowledge-driven reasoning core. Our key technical innovation is a sophisticated fusion module, inspired by the Q-Former architecture, which distills complex spatiotemporal and object-centric visual features into a concise, language-aligned representation. This enables the LLM to effectively ground its inferential processes in direct visual evidence. The model is trained via a two-stage strategy, beginning with large-scale alignment pre-training on video-text data, followed by targeted instruction fine-tuning on a curated dataset designed to elicit advanced reasoning and prediction skills. Extensive experiments demonstrate that our model achieves state-of-the-art performance on multiple challenging benchmarks. Notably, it exhibits remarkable zero-shot generalization to unseen reasoning tasks, and our in-depth ablation studies validate the critical contribution of each architectural component. This work pushes the boundary of machine perception from simple recognition towards genuine cognitive understanding, paving the way for more intelligent and capable AI systems in robotics, human-computer interaction, and beyond.
Authors: Wenkang Zhang, Yan Zhao, Qiang Wang, Zhixin Xu, Li Song, Zhengxue Cheng
Abstract: Free-Viewpoint Video (FVV) enables immersive 3D experiences, but efficient compression of dynamic 3D representation remains a major challenge. Existing dynamic 3D Gaussian Splatting methods couple reconstruction with optimization-dependent compression and customized motion formats, limiting generalization and standardization. To address this, we propose D-FCGS, a novel Feedforward Compression framework for Dynamic Gaussian Splatting. Key innovations include: (1) a standardized Group-of-Frames (GoF) structure with I-P coding, leveraging sparse control points to extract inter-frame motion tensors; (2) a dual prior-aware entropy model that fuses hyperprior and spatial-temporal priors for accurate rate estimation; (3) a control-point-guided motion compensation mechanism and refinement network to enhance view-consistent fidelity. Trained on Gaussian frames derived from multi-view videos, D-FCGS generalizes across diverse scenes in a zero-shot fashion. Experiments show that it matches the rate-distortion performance of optimization-based methods, achieving over 17 times compression compared to the baseline while preserving visual quality across viewpoints. This work advances feedforward compression of dynamic 3DGS, facilitating scalable FVV transmission and storage for immersive applications.
Authors: Heitor R. Medeiros, Atif Belal, Masih Aminbeidokhti, Eric Granger, Marco Pedersoli
Abstract: Object detection (OD) in infrared (IR) imagery is critical for low-light and nighttime applications. However, the scarcity of large-scale IR datasets forces models to rely on weights pre-trained on RGB images. While fine-tuning on IR improves accuracy, it often compromises robustness under distribution shifts due to the inherent modality gap between RGB and IR. To address this, we introduce LLVIP-C and FLIR-C, two cross-modality out-of-distribution (OOD) benchmarks built by applying corruptions to standard IR datasets. Additionally, to fully leverage the complementary knowledge from RGB and infrared-trained models, we propose WiSE-OD, a weight-space ensembling method with two variants: WiSE-OD$_{ZS}$, which combines RGB zero-shot and IR fine-tuned weights, and WiSE-OD$_{LP}$, which blends zero-shot and linear probing. Evaluated using four RGB-pretrained detectors and two robust baselines on our benchmark and in the real-world out-of-distribution M3FD dataset, our WiSE-OD improves robustness across modalities and to corruption in synthetic and real-world distribution shifts without any additional training or inference costs. Our code is available at: https://github.com/heitorrapela/wiseod.
Authors: Yan Li, Ze Yang, Keisuke Tateno, Federico Tombari, Liang Zhao, Gim Hee Lee
Abstract: Minimal parametrization of 3D lines plays a critical role in camera localization and structural mapping. Existing representations in robotics and computer vision predominantly handle independent lines, overlooking structural regularities such as sets of parallel lines that are pervasive in man-made environments. This paper introduces \textbf{RiemanLine}, a unified minimal representation for 3D lines formulated on Riemannian manifolds that jointly accommodates both individual lines and parallel-line groups. Our key idea is to decouple each line landmark into global and local components: a shared vanishing direction optimized on the unit sphere $\mathcal{S}^2$, and scaled normal vectors constrained on orthogonal subspaces, enabling compact encoding of structural regularities. For $n$ parallel lines, the proposed representation reduces the parameter space from $4n$ (orthonormal form) to $2n+2$, naturally embedding parallelism without explicit constraints. We further integrate this parameterization into a factor graph framework, allowing global direction alignment and local reprojection optimization within a unified manifold-based bundle adjustment. Extensive experiments on ICL-NUIM, TartanAir, and synthetic benchmarks demonstrate that our method achieves significantly more accurate pose estimation and line reconstruction, while reducing parameter dimensionality and improving convergence stability.
Authors: Haoyu Liu, Chaoyu Gong, Mengke He, Jiate Li, Kai Han, Siqiang Luo
Abstract: The proliferation of generative video models has made detecting AI-generated and manipulated videos an urgent challenge. Existing detection approaches often fail to generalize across diverse manipulation types due to their reliance on isolated spatial, temporal, or spectral information, and typically require large models to perform well. This paper introduces SSTGNN, a lightweight Spatial-Spectral-Temporal Graph Neural Network framework that represents videos as structured graphs, enabling joint reasoning over spatial inconsistencies, temporal artifacts, and spectral distortions. SSTGNN incorporates learnable spectral filters and spatial-temporal differential modeling into a unified graph-based architecture, capturing subtle manipulation traces more effectively. Extensive experiments on diverse benchmark datasets demonstrate that SSTGNN not only achieves superior performance in both in-domain and cross-domain settings, but also offers strong efficiency and resource allocation. Remarkably, SSTGNN accomplishes these results with up to 42$\times$ fewer parameters than state-of-the-art models, making it highly lightweight and resource-friendly for real-world deployment.
Authors: Yuxin Mao, Zhen Qin, Jinxing Zhou, Bin Fan, Jing Zhang, Yiran Zhong, Yuchao Dai
Abstract: Vision Transformers (ViTs) have revolutionized computer vision, yet their self-attention mechanism lacks explicit spatial inductive biases, leading to suboptimal performance on spatially-structured tasks. Existing approaches introduce data-independent spatial decay based on fixed distance metrics, applying uniform attention weighting regardless of image content and limiting adaptability to diverse visual scenarios. Inspired by recent advances in large language models where content-aware gating mechanisms (e.g., GLA, HGRN2, FOX) significantly outperform static alternatives, we present the first successful adaptation of data-dependent spatial decay to 2D vision transformers. We introduce \textbf{Spatial Decay Transformer (SDT)}, featuring a novel Context-Aware Gating (CAG) mechanism that generates dynamic, data-dependent decay for patch interactions. Our approach learns to modulate spatial attention based on both content relevance and spatial proximity. We address the fundamental challenge of 1D-to-2D adaptation through a unified spatial-content fusion framework that integrates manhattan distance-based spatial priors with learned content representations. Extensive experiments on ImageNet-1K classification and generation tasks demonstrate consistent improvements over strong baselines. Our work establishes data-dependent spatial decay as a new paradigm for enhancing spatial attention in vision transformers.
Authors: Zhenyi Zhao, Muthu Rama Krishnan Mookiah, Emanuele Trucco
Abstract: Purpose: This study aims to introduce the first adaptation of RETFound for joint optic disc (OD) and optic cup (OC) segmentation. RETFound is a well-known foundation model developed for fundus camera and optical coherence tomography images, which has shown promising performance in disease diagnosis. Methods: We propose FunduSegmenter, a model integrating a series of novel modules with RETFound, including a Pre-adapter, a Decoder, a Post-adapter, skip connections with Convolutional Block Attention Module and a Vision Transformer block adapter. The model is evaluated on a private dataset, GoDARTS, and four public datasets, IDRiD, Drishti-GS, RIM-ONE-r3, and REFUGE, through internal verification, external verification and domain generalization experiments. Results: An average Dice similarity coefficient of 90.51% was achieved in internal verification, which substantially outperformed the baselines (nnU-Net: 82.91%; DUNet: 89.17%; TransUNet: 87.91%). In all external verification experiments, the average results were about 3% higher than those of the best baseline, and were also competitive in domain generalization. Conclusions: This study explored the potential of the latent general representations learned by RETFound for OD and OC segmentation in fundus camera images. Our FunduSegmenter outperformed nearly all state-of-the-art baseline methods. The proposed modules are general and can be extended to fine-tuning other foundation models. Translational Relevance: The model shows strong stability and generalization on both in-distribution and out-of-distribution data, providing stable OD and OC segmentation. This is an essential step for many automated tasks, from setting the accurate retinal coordinate to biomarker discovery. The code and all trained weights are available at: [link to be added after the paper is accepted]
Authors: Vipooshan Vipulananthan, Kumudu Mohottala, Kavindu Chinthana, Nimsara Paramulla, Charith D Chitraranjan
Abstract: Accident prediction and timely preventive actions improve road safety by reducing the risk of injury to road users and minimizing property damage. Hence, they are critical components of advanced driver assistance systems (ADAS) and autonomous vehicles. While many existing systems depend on multiple sensors such as LiDAR, radar, and GPS, relying solely on dash-cam videos presents a more challenging, yet more cost-effective and easily deployable solution. In this work, we incorporate improved spatio-temporal features and aggregate them through a recurrent network to enhance state-of-the-art graph neural networks for predicting accidents from dash-cam videos. Experiments using three publicly available datasets (DAD, DoTA and DADA) show that our proposed STAGNet model achieves higher average precision and mean time-to-accident scores than previous methods, both when cross-validated on a given dataset and when trained and tested on different datasets.
Authors: Zhongmiao Qi, Yan Jiang, Bolin Zhang, Chong Wang, Lijun Guo, Pengjiang Qian, Jiangbo Qian
Abstract: With the explosive growth of video data in various complex scenarios, quickly retrieving group activities has become an urgent problem. However, many tasks can only retrieve videos focusing on an entire video, not the activity granularity. To solve this problem, we propose a new STVH (spatiotemporal interleaved video hashing) technique for the first time. Through a unified framework, the STVH simultaneously models individual object dynamics and group interactions, capturing the spatiotemporal evolution on both group visual features and positional features. Moreover, in real-life video retrieval scenarios, it may sometimes require activity features, while at other times, it may require visual features of objects. We then further propose a novel M-STVH (multi-focused spatiotemporal video hashing) as an enhanced version to handle this difficult task. The advanced method incorporates hierarchical feature integration through multi-focused representation learning, allowing the model to jointly focus on activity semantics features and object visual features. We conducted comparative experiments on publicly available datasets, and both STVH and M-STVH can achieve excellent results.
Authors: Alireza Sedighi Moghaddam, Mohammad Reza Mohammadi
Abstract: Labeled data is a fundamental component in training supervised deep learning models for computer vision tasks. However, the labeling process, especially for ordinal image classification where class boundaries are often ambiguous, is prone to error and noise. Such label noise can significantly degrade the performance and reliability of machine learning models. This paper addresses the problem of detecting and correcting label noise in ordinal image classification tasks. To this end, a novel data-centric method called ORDinal Adaptive Correction (ORDAC) is proposed for adaptive correction of noisy labels. The proposed approach leverages the capabilities of Label Distribution Learning (LDL) to model the inherent ambiguity and uncertainty present in ordinal labels. During training, ORDAC dynamically adjusts the mean and standard deviation of the label distribution for each sample. Rather than discarding potentially noisy samples, this approach aims to correct them and make optimal use of the entire training dataset. The effectiveness of the proposed method is evaluated on benchmark datasets for age estimation (Adience) and disease severity detection (Diabetic Retinopathy) under various asymmetric Gaussian noise scenarios. Results show that ORDAC and its extended versions (ORDAC_C and ORDAC_R) lead to significant improvements in model performance. For instance, on the Adience dataset with 40% noise, ORDAC_R reduced the mean absolute error from 0.86 to 0.62 and increased the recall metric from 0.37 to 0.49. The method also demonstrated its effectiveness in correcting intrinsic noise present in the original datasets. This research indicates that adaptive label correction using label distributions is an effective strategy to enhance the robustness and accuracy of ordinal classification models in the presence of noisy data.
Authors: Hao Yin, Xin Man, Feiyu Chen, Jie Shao, Heng Tao Shen
Abstract: Text-to-Image Person Retrieval (TIPR) is a cross-modal matching task designed to identify the person images that best correspond to a given textual description. The key difficulty in TIPR is to realize robust correspondence between the textual and visual modalities within a unified latent representation space. To address this challenge, prior approaches incorporate attention mechanisms for implicit cross-modal local alignment. However, they lack the ability to verify whether all local features are correctly aligned. Moreover, existing methods tend to emphasize the utilization of hard negative samples during model optimization to strengthen discrimination between positive and negative pairs, often neglecting incorrectly matched positive pairs. To mitigate these problems, we propose FMFA, a cross-modal Full-Mode Fine-grained Alignment framework, which enhances global matching through explicit fine-grained alignment and existing implicit relational reasoning -- hence the term ``full-mode'' -- without introducing extra supervisory signals. In particular, we propose an Adaptive Similarity Distribution Matching (A-SDM) module to rectify unmatched positive sample pairs. A-SDM adaptively pulls the unmatched positive pairs closer in the joint embedding space, thereby achieving more precise global alignment. Additionally, we introduce an Explicit Fine-grained Alignment (EFA) module, which makes up for the lack of verification capability of implicit relational reasoning. EFA strengthens explicit cross-modal fine-grained interactions by sparsifying the similarity matrix and employs a hard coding method for local alignment. We evaluate our method on three public datasets, where it attains state-of-the-art results among all global matching methods. The code for our method is publicly accessible at https://github.com/yinhao1102/FMFA.
Authors: Antonio Scardace, Lemuel Puglisi, Francesco Guarnera, Sebastiano Battiato, Daniele Rav\`i
Abstract: Deep generative models have emerged as a transformative tool in medical imaging, offering substantial potential for synthetic data generation. However, recent empirical studies highlight a critical vulnerability: these models can memorize sensitive training data, posing significant risks of unauthorized patient information disclosure. Detecting memorization in generative models remains particularly challenging, necessitating scalable methods capable of identifying training data leakage across large sets of generated samples. In this work, we propose DeepSSIM, a novel self-supervised metric for quantifying memorization in generative models. DeepSSIM is trained to: i) project images into a learned embedding space and ii) force the cosine similarity between embeddings to match the ground-truth SSIM (Structural Similarity Index) scores computed in the image space. To capture domain-specific anatomical features, training incorporates structure-preserving augmentations, allowing DeepSSIM to estimate similarity reliably without requiring precise spatial alignment. We evaluate DeepSSIM in a case study involving synthetic brain MRI data generated by a Latent Diffusion Model (LDM) trained under memorization-prone conditions, using 2,195 MRI scans from two publicly available datasets (IXI and CoRR). Compared to state-of-the-art memorization metrics, DeepSSIM achieves superior performance, improving F1 scores by an average of +52.03% over the best existing method. Code and data of our approach are publicly available at the following link: https://github.com/brAIn-science/DeepSSIM.
Authors: Roie Kazoom, Alon Goldberg, Hodaya Cohen, Ofer Hadar
Abstract: Adversarial patch attacks pose a severe threat to deep neural networks, yet most existing approaches rely on unrealistic white-box assumptions, untargeted objectives, or produce visually conspicuous patches that limit real-world applicability. In this work, we introduce a novel framework for fully controllable adversarial patch generation, where the attacker can freely choose both the input image x and the target class y target, thereby dictating the exact misclassification outcome. Our method combines a generative U-Net design with Grad-CAM-guided patch placement, enabling semantic-aware localization that maximizes attack effectiveness while preserving visual realism. Extensive experiments across convolutional networks (DenseNet-121, ResNet-50) and vision transformers (ViT-B/16, Swin-B/16, among others) demonstrate that our approach achieves state-of-the-art performance across all settings, with attack success rates (ASR) and target-class success (TCS) consistently exceeding 99%. Importantly, we show that our method not only outperforms prior white-box attacks and untargeted baselines, but also surpasses existing non-realistic approaches that produce detectable artifacts. By simultaneously ensuring realism, targeted control, and black-box applicability-the three most challenging dimensions of patch-based attacks-our framework establishes a new benchmark for adversarial robustness research, bridging the gap between theoretical attack strength and practical stealthiness.
Authors: Nate Rothschild, Moshe Kimhi, Avi Mendelson, Chaim Baskin
Abstract: Image reconstruction from corrupted images is crucial across many domains. Most reconstruction networks are trained on post-ISP sRGB images, even though the image-signal-processing pipeline irreversibly mixes colors, clips dynamic range, and blurs fine detail. This paper uses the rain degradation problem as a use case to show that these losses are avoidable, and demonstrates that learning directly on raw Bayer mosaics yields superior reconstructions. To substantiate the claim, we (i) evaluate post-ISP and Bayer reconstruction pipelines, (ii) curate Raw-Rain, the first public benchmark of real rainy scenes captured in both 12-bit Bayer and bit-depth-matched sRGB, and (iii) introduce Information Conservation Score (ICS), a color-invariant metric that aligns more closely with human opinion than PSNR or SSIM. On the test split, our raw-domain model improves sRGB results by up to +0.99 dB PSNR and +1.2% ICS, while running faster with half of the GFLOPs. The results advocate an ISP-last paradigm for low-level vision and open the door to end-to-end learnable camera pipelines.
Authors: KunHo Heo, GiHyun Kim, SuYeon Kim, MyeongAh Cho
Abstract: 3D Semantic Scene Graph Prediction aims to detect objects and their semantic relationships in 3D scenes, and has emerged as a crucial technology for robotics and AR/VR applications. While previous research has addressed dataset limitations and explored various approaches including Open-Vocabulary settings, they frequently fail to optimize the representational capacity of object and relationship features, showing excessive reliance on Graph Neural Networks despite insufficient discriminative capability. In this work, we demonstrate through extensive analysis that the quality of object features plays a critical role in determining overall scene graph accuracy. To address this challenge, we design a highly discriminative object feature encoder and employ a contrastive pretraining strategy that decouples object representation learning from the scene graph prediction. This design not only enhances object classification accuracy but also yields direct improvements in relationship prediction. Notably, when plugging in our pretrained encoder into existing frameworks, we observe substantial performance improvements across all evaluation metrics. Additionally, whereas existing approaches have not fully exploited the integration of relationship information, we effectively combine both geometric and semantic features to achieve superior relationship prediction. Comprehensive experiments on the 3DSSG dataset demonstrate that our approach significantly outperforms previous state-of-the-art methods. Our code is publicly available at https://github.com/VisualScienceLab-KHU/OCRL-3DSSG-Codes.
URLs: https://github.com/VisualScienceLab-KHU/OCRL-3DSSG-Codes.
Authors: Zeteng Lin, Xingxing Li, Wen You, Xiaoyang Li, Zehan Lu, Yujun Cai, Jing Tang
Abstract: Existing vision language models (VLMs), including GPT-4 and DALL.E, often struggle to preserve logic, object identity, and style in multimodal image-text generation. This limitation significantly hinders the generalization capability of VLMs in complex image-text input-output scenarios. To address this issue, we propose IUT-Plug, a module grounded in an Image Understanding Tree (IUT), which enhances existing interleaved VLMs through explicit structured reasoning, thereby mitigating context drift in logic, entity identity, and style. The proposed framework operates in two stages. (1) A dynamic IUT-Plug extraction module parses visual scenes into hierarchical symbolic structures. (2) A coordinated narrative-flow and image synthesis mechanism ensures cross-modal consistency. To evaluate our approach, we construct a novel benchmark based on 3,000 real human-generated question-answer pairs over fine-tuned large models, introducing a dynamic evaluation protocol for quantifying context drift in interleaved VLMs. Experimental results demonstrate that IUT-Plug not only improves accuracy on established benchmarks but also effectively alleviates the three critical forms of context drift across diverse multimodal question answering (QA) scenarios.
Authors: Frangil Ramirez, Elizabeth Clerkin, David J. Crandall, Linda B. Smith
Abstract: One-year-old infants show immediate adult-like generalization of common object categories to novel instances. The field has limited understanding of how this early prowess is achieved. Here we provide evidence on infants' daily-life visual experiences for 8 early-learned object categories. Using a corpus of infant head camera images recorded at mealtimes (87 mealtimes captured by 14 infants), we quantify the individual instances experienced by infants and the similarity structure of all images containing an instance of each category. The distribution of instances is highly skewed, containing, for each infant and category, many images of the same few objects along with fewer images of other instances. Graph theoretic measures of the similarity structure for individual categories reveal a lumpy mix of high similarity and high variability, organized into multiple but interconnected clusters of high-similarity images. In computational experiments, we show that creating training sets that include an oversampling of varied images from a single instance yields a lumpy similarity structure. We also show that these artificially-created training sets support generalization to novel instances after very few training experiences. We discuss implications for the development of visual object recognition in both humans and machines.
Authors: Weijie Wang, Jiagang Zhu, Zeyu Zhang, Xiaofeng Wang, Zheng Zhu, Guosheng Zhao, Chaojun Ni, Haoxiao Wang, Guan Huang, Xinze Chen, Yukun Zhou, Wenkang Qin, Duochao Shi, Haoyun Li, Yicheng Xiao, Donny Y. Chen, Jiwen Lu
Abstract: We present DriveGen3D, a novel framework for generating high-quality and highly controllable dynamic 3D driving scenes that addresses critical limitations in existing methodologies. Current approaches to driving scene synthesis either suffer from prohibitive computational demands for extended temporal generation, focus exclusively on prolonged video synthesis without 3D representation, or restrict themselves to static single-scene reconstruction. Our work bridges this methodological gap by integrating accelerated long-term video generation with large-scale dynamic scene reconstruction through multimodal conditional control. DriveGen3D introduces a unified pipeline consisting of two specialized components: FastDrive-DiT, an efficient video diffusion transformer for high-resolution, temporally coherent video synthesis under text and Bird's-Eye-View (BEV) layout guidance; and FastRecon3D, a feed-forward module that rapidly builds 3D Gaussian representations across time, ensuring spatial-temporal consistency. DriveGen3D enable the generation of long driving videos (up to $800\times424$ at $12$ FPS) and corresponding 3D scenes, achieving state-of-the-art results while maintaining efficiency.
Authors: Mhd Adnan Albani, Riad Sonbol
Abstract: Parkinson's disease (PD) is a neurodegenerative disease affecting about 1% of people over the age of 60, causing motor impairments that impede hand coordination activities such as writing and drawing. Many approaches have tried to support early detection of Parkinson's disease based on hand-drawn images; however, we identified two major limitations in the related works: (1) the lack of sufficient datasets, (2) the robustness when dealing with unseen patient data. In this paper, we propose a new approach to detect Parkinson's disease that consists of two stages: The first stage classifies based on their drawing type(circle, meander, spiral), and the second stage extracts the required features from the images and detects Parkinson's disease. We overcame the previous two limitations by applying a chunking strategy where we divide each image into 2x2 chunks. Each chunk is processed separately when extracting features and recognizing Parkinson's disease indicators. To make the final classification, an ensemble method is used to merge the decisions made from each chunk. Our evaluation shows that our proposed approach outperforms the top performing state-of-the-art approaches, in particular on unseen patients. On the NewHandPD dataset our approach, it achieved 97.08% accuracy for seen patients and 94.91% for unseen patients, our proposed approach maintained a gap of only 2.17 percentage points, compared to the 4.76-point drop observed in prior work.
Authors: Zhiqiang Teng, Tingting Chen, Beibei Lin, Zifeng Yuan, Xuanyi Li, Xuanyu Zhang, Shunli Zhang
Abstract: 3D Gaussian Splatting (3DGS) under raindrop conditions suffers from severe occlusions and optical distortions caused by raindrop contamination on the camera lens, substantially degrading reconstruction quality. Existing benchmarks typically evaluate 3DGS using synthetic raindrop images with known camera poses (constrained images), assuming ideal conditions. However, in real-world scenarios, raindrops often interfere with accurate camera pose estimation and point cloud initialization. Moreover, a significant domain gap between synthetic and real raindrops further impairs generalization. To tackle these issues, we introduce RaindropGS, a comprehensive benchmark designed to evaluate the full 3DGS pipeline-from unconstrained, raindrop-corrupted images to clear 3DGS reconstructions. Specifically, the whole benchmark pipeline consists of three parts: data preparation, data processing, and raindrop-aware 3DGS evaluation, including types of raindrop interference, camera pose estimation and point cloud initialization, single image rain removal comparison, and 3D Gaussian training comparison. First, we collect a real-world raindrop reconstruction dataset, in which each scene contains three aligned image sets: raindrop-focused, background-focused, and rain-free ground truth, enabling a comprehensive evaluation of reconstruction quality under different focus conditions. Through comprehensive experiments and analyses, we reveal critical insights into the performance limitations of existing 3DGS methods on unconstrained raindrop images and the varying impact of different pipeline components: the impact of camera focus position on 3DGS reconstruction performance, and the interference caused by inaccurate pose and point cloud initialization on reconstruction. These insights establish clear directions for developing more robust 3DGS methods under raindrop conditions.
Authors: Moona Mazher, Geoff J. M. Parker, Daniel C. Alexander
Abstract: Foundation models in artificial intelligence (AI) are transforming medical imaging by enabling general-purpose feature learning from large-scale, unlabeled datasets. In this work, we introduce BrainFound, a self-supervised foundation model for brain MRI, built by extending DINO-v2, a vision transformer originally designed for 2D natural images. BrainFound adapts DINO-v2 to model full 3D brain anatomy by incorporating volumetric information from sequential MRI slices, moving beyond conventional single-slice paradigms. It supports both single- and multimodal inputs, enabling a broad range of downstream tasks, including disease detection and image segmentation, while generalising across varied imaging protocols and clinical scenarios. We show that BrainFound consistently outperforms existing self-supervised pretraining strategies and supervised baselines, particularly in label-scarce and multi-contrast settings. By integrating information from diverse 3D MRI modalities (e.g., T1, T2, FLAIR), it enhances diagnostic accuracy and reduces dependency on extensive expert annotations. This flexibility makes BrainFound a scalable and practical solution for 3D neuroimaging pipelines, with significant potential for clinical deployment and research innovation.
Authors: Xin Jin, Siyuan Li, Siyong Jian, Kai Yu, Huan Wang
Abstract: Vision-language alignment in multi-modal large language models (MLLMs) relies on supervised fine-tuning (SFT) or reinforcement learning (RL). To align multi-modal large language models (MLLMs) in the post-training stage, supervised fine-tuning (SFT) is a stable choice but requires human annotations and lacks task generalizations, while Reinforcement Learning (RL) searches for better answers from reward signals but suffers from computational overhead and instability. To achieve balance among scalability, efficiency, and alignment generalizations, we propose MergeMix, a unified paradigm that bridges SFT and RL with an efficient Token Merge based Mixup augmentation. As for the Mixup policy, we generate contextual aligned mixed images with the corresponding labels according to the merged attention maps with cluster regions. Then, we enhance the preference-driven paradigm for MLLMs by building preference pairs with raw images and MergeMix-generated ones and optimizing the soft preference margin with the mixed SimPO loss. Extensive experiments demonstrate that MergeMix not only achieves dominant classification accuracy as an augmentation method but also improves generalization abilities and alignment of MLLMs, providing a new learning paradigm for preference alignment with training efficiency and stability.
Authors: Shengqian Zhu, Chengrong Yu, Qiang Wang, Ying Song, Guangjun Li, Jiafei Wu, Xiaogang Xu, Zhang Yi, Junjie Hu
Abstract: Class incremental medical image segmentation (CIMIS) aims to preserve knowledge of previously learned classes while learning new ones without relying on old-class labels. However, existing methods 1) either adopt one-size-fits-all strategies that treat all spatial regions and feature channels equally, which may hinder the preservation of accurate old knowledge, 2) or focus solely on aligning local prototypes with global ones for old classes while overlooking their local representations in new data, leading to knowledge degradation. To mitigate the above issues, we propose Prototype-Guided Calibration Distillation (PGCD) and Dual-Aligned Prototype Distillation (DAPD) for CIMIS in this paper. Specifically, PGCD exploits prototype-to-feature similarity to calibrate class-specific distillation intensity in different spatial regions, effectively reinforcing reliable old knowledge and suppressing misleading information from old classes. Complementarily, DAPD aligns the local prototypes of old classes extracted from the current model with both global prototypes and local prototypes, further enhancing segmentation performance on old categories. Comprehensive evaluations on two widely used multi-organ segmentation benchmarks demonstrate that our method outperforms state-of-the-art methods, highlighting its robustness and generalization capabilities.
Authors: Zhicheng Liao, Dongxu Wu, Zhenshan Shi, Sijie Mai, Hanwei Zhu, Lingyu Zhu, Yuncheng Jiang, Baoliang Chen
Abstract: Recent efforts have repurposed the Contrastive Language-Image Pre-training (CLIP) model for No-Reference Image Quality Assessment (NR-IQA) by measuring the cosine similarity between the image embedding and textual prompts such as "a good photo" or "a bad photo." However, this semantic similarity overlooks a critical yet underexplored cue: the magnitude of the CLIP image features, which we empirically find to exhibit a strong correlation with perceptual quality. In this work, we introduce a novel adaptive fusion framework that complements cosine similarity with a magnitude-aware quality cue. Specifically, we first extract the absolute CLIP image features and apply a Box-Cox transformation to statistically normalize the feature distribution and mitigate semantic sensitivity. The resulting scalar summary serves as a semantically-normalized auxiliary cue that complements cosine-based prompt matching. To integrate both cues effectively, we further design a confidence-guided fusion scheme that adaptively weighs each term according to its relative strength. Extensive experiments on multiple benchmark IQA datasets demonstrate that our method consistently outperforms standard CLIP-based IQA and state-of-the-art baselines, without any task-specific training.
Authors: Zheyuan Zhang, Jiwei Zhang, Boyu Zhou, Linzhimeng Duan, Hong Chen
Abstract: Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2's exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2% (compared to CricaVPR).Code is available at https://github.com/tony19980810/D2VPR.
Authors: Yoonjae Seo, Ermal Elbasani, Jaehong Lee
Abstract: Most neural network quantization methods apply uniform bit precision across spatial regions, disregarding the heterogeneous complexity inherent in visual data. This paper introduces MCAQ-YOLO, a practical framework for tile-wise spatial mixed-precision quantization in real-time object detectors. Morphological complexity--quantified through five complementary metrics (fractal dimension, texture entropy, gradient variance, edge density, and contour complexity)--is proposed as a signal-centric predictor of spatial quantization sensitivity. A calibration-time analysis design enables spatial bit allocation with only 0.3ms inference overhead, achieving 151 FPS throughput. Additionally, a curriculum-based training scheme that progressively increases quantization difficulty is introduced to stabilize optimization and accelerate convergence. On a construction safety equipment dataset exhibiting high morphological variability, MCAQ-YOLO achieves 85.6% mAP@0.5 with an average bit-width of 4.2 bits and a 7.6x compression ratio, outperforming uniform 4-bit quantization by 3.5 percentage points. Cross-dataset evaluation on COCO 2017 (+2.9%) and Pascal VOC 2012 (+2.3%) demonstrates consistent improvements, with performance gains correlating with within-image complexity variation.
Authors: Junhee Lee, ChaeBeen Bang, MyoungChul Kim, MyeongAh Cho
Abstract: Weakly-Supervised Video Anomaly Detection aims to identify anomalous events using only video-level labels, balancing annotation efficiency with practical applicability. However, existing methods often oversimplify the anomaly space by treating all abnormal events as a single category, overlooking the diverse semantic and temporal characteristics intrinsic to real-world anomalies. Inspired by how humans perceive anomalies, by jointly interpreting temporal motion patterns and semantic structures underlying different anomaly types, we propose RefineVAD, a novel framework that mimics this dual-process reasoning. Our framework integrates two core modules. The first, Motion-aware Temporal Attention and Recalibration (MoTAR), estimates motion salience and dynamically adjusts temporal focus via shift-based attention and global Transformer-based modeling. The second, Category-Oriented Refinement (CORE), injects soft anomaly category priors into the representation space by aligning segment-level features with learnable category prototypes through cross-attention. By jointly leveraging temporal dynamics and semantic structure, explicitly models both "how" motion evolves and "what" semantic category it resembles. Extensive experiments on WVAD benchmark validate the effectiveness of RefineVAD and highlight the importance of integrating semantic context to guide feature refinement toward anomaly-relevant patterns.
Authors: Yuanchao Wang, Tian Qin, Eduardo Valle, Bruno Abrahao
Abstract: Out-of-distribution (OOD) detection is critical for deploying image classifiers in safety-sensitive environments, yet existing detectors often struggle when OOD samples are semantically similar to the in-distribution (ID) classes. We present BootOOD, a fully self-supervised OOD detection framework that bootstraps exclusively from ID data and is explicitly designed to handle semantically challenging OOD samples. BootOOD synthesizes pseudo-OOD features through simple transformations of ID representations and leverages Neural Collapse (NC), where ID features cluster tightly around class means with consistent feature norms. Unlike prior approaches that aim to constrain OOD features into subspaces orthogonal to the collapsed ID means, BootOOD introduces a lightweight auxiliary head that performs radius-based classification on feature norms. This design decouples OOD detection from the primary classifier and imposes a relaxed requirement: OOD samples are learned to have smaller feature norms than ID features, which is easier to satisfy when ID and OOD are semantically close. Experiments on CIFAR-10, CIFAR-100, and ImageNet-200 show that BootOOD outperforms prior post-hoc methods, surpasses training-based methods without outlier exposure, and is competitive with state-of-the-art outlier-exposure approaches while maintaining or improving ID accuracy.
Authors: Zhonghao Liu, Hanxue Gu, Qihang Li, Michael Fox, Jay M. Levin, Maciej A. Mazurowski, Brian C. Lau
Abstract: To develop and validate a fully automated, deep-learning pipeline for measuring glenoid bone loss on 3D CT scans using linear-based, en-face view, and best-circle method. Shoulder CT scans of 81 patients were retrospectively collected between January 2013 and March 2023. Our algorithm consists of three main stages: (1) Segmentation, where we developed a U-Net to automatically segment the glenoid and humerus; (2) anatomical landmark detection, where a second network predicts glenoid rim points; and (3) geometric fitting, where we applied a principal component analysis (PCA), projection, and circle fitting to compute the percentage of bone loss. The performance of the pipeline was evaluated using DSC for segmentation and MAE and ICC for bone-loss measurement; intermediate outputs (rim point sets and en-face view) were also assessed. Automated measurements showed strong agreement with consensus readings, exceeding surgeon-to-surgeon consistency (ICC 0.84 vs 0.78 for all patients; ICC 0.71 vs 0.63 for low bone loss; ICC 0.83 vs 0.21 for high bone loss; P < 0.001). For the classification task of assigning each patient to different bone loss severity subgroups, the pipeline's sensitivity was 71.4% for the low-severity group and 85.7% for the high-severity group, with no instances of misclassifying low as high or vice versa. A fully automated, deep learning-based pipeline for glenoid bone-loss measurement on CT scans can be a clinically reliable tool to assist clinicians with preoperative planning for shoulder instability. We are releasing our model and dataset at https://github.com/Edenliu1/Auto-Glenoid-Measurement-DL-Pipeline .
URLs: https://github.com/Edenliu1/Auto-Glenoid-Measurement-DL-Pipeline
Authors: Shanshan Zhang, Liqin Wu, Wenying Cao, Siyue Wang, Tianshui Wen, Qi Zhang, Xuemin Hong, Ao Peng, Lingxiang Zheng, Yu Yang
Abstract: Inertial Odometry (IO) enables real-time localization using only acceleration and angular velocity measurements from an Inertial Measurement Unit (IMU), making it a promising solution for localization in consumer-grade applications. Traditionally, researchers have routinely transformed IMU measurements into the global frame to obtain smoother motion representations. However, recent studies in drone scenarios have demonstrated that the body frame can significantly improve localization accuracy, prompting a re-evaluation of the suitability of the global frame for pedestrian IO. To address this issue, this paper systematically evaluates the effectiveness of the global frame in pedestrian IO through theoretical analysis, qualitative inspection, and quantitative experiments. Building upon these findings, we further propose MambaIO, which decomposes IMU measurements into high-frequency and low-frequency components using a Laplacian pyramid. The low-frequency component is processed by a Mamba architecture to extract implicit contextual motion cues, while the high-frequency component is handled by a convolutional structure to capture fine-grained local motion details. Experiments on multiple public datasets show that MambaIO substantially reduces localization error and achieves state-of-the-art (SOTA) performance. To the best of our knowledge, this is the first application of the Mamba architecture to the IO task.
Authors: Zhimin Shao, Abhay Yadav, Rama Chellappa, Cheng Peng
Abstract: Reliable image correspondences form the foundation of vision-based spatial perception, enabling recovery of 3D structure and camera poses. However, unconstrained feature matching across domains such as aerial, indoor, and outdoor scenes remains challenging due to large variations in appearance, scale and viewpoint. Feature matching has been conventionally formulated as a 2D-to-2D problem; however, recent 3D foundation models provides spatial feature matching properties based on two-view geometry. While powerful, we observe that these spatially coherent matches often concentrate on dominant planar regions, e.g., walls or ground surfaces, while being less sensitive to fine-grained geometric details, particularly under large viewpoint changes. To better understand these trade-offs, we first perform linear probe experiments to evaluate the performance of various vision foundation models for image matching. Building on these insights, we introduce SPIDER, a universal feature matching framework that integrates a shared feature extraction backbone with two specialized network heads for estimating both 2D-based and 3D-based correspondences from coarse to fine. Finally, we introduce an image-matching evaluation benchmark that focuses on unconstrained scenarios with large baselines. SPIDER significantly outperforms SoTA methods, demonstrating its strong ability as a universal image-matching method.
Authors: Wenhao Guo, Golrokh Mirzaei
Abstract: Differentiating true tumor progression (TP) from treatment-related pseudoprogression (PsP) in glioblastoma remains challenging, especially at early follow-up. We present the first stage-specific, cross-sectional benchmarking of deep learning models for follow-up MRI using the Burdenko GBM Progression cohort (n = 180). We analyze different post-RT scans independently to test whether architecture performance depends on time-point. Eleven representative DL families (CNNs, LSTMs, hybrids, transformers, and selective state-space models) were trained under a unified, QC-driven pipeline with patient-level cross-validation. Across both stages, accuracies were comparable (~0.70-0.74), but discrimination improved at the second follow-up, with F1 and AUC increasing for several models, indicating richer separability later in the care pathway. A Mamba+CNN hybrid consistently offered the best accuracy-efficiency trade-off, while transformer variants delivered competitive AUCs at substantially higher computational cost and lightweight CNNs were efficient but less reliable. Performance also showed sensitivity to batch size, underscoring the need for standardized training protocols. Notably, absolute discrimination remained modest overall, reflecting the intrinsic difficulty of TP vs. PsP and the dataset's size imbalance. These results establish a stage-aware benchmark and motivate future work incorporating longitudinal modeling, multi-sequence MRI, and larger multi-center cohorts.
Authors: Samuele Dell'Erba, Andrew D. Bagdanov
Abstract: Diffusion models have established the state-of-the-art in text-to-image generation, but their performance often relies on a diffusion prior network to translate text embeddings into the visual manifold for easier decoding. These priors are computationally expensive and require extensive training on massive datasets. In this work, we challenge the necessity of a trained prior at all by employing Optimization-based Visual Inversion (OVI), a training-free and zero-shot alternative, to replace the need for a prior. OVI initializes a latent visual representation from random pseudo-tokens and iteratively optimizes it to maximize the cosine similarity with the input textual prompt embedding. We further propose two novel constraints, a Mahalanobis-based and a Nearest-Neighbor loss, to regularize the OVI optimization process toward the distribution of realistic images. Our experiments, conducted on Kandinsky 2.2, show that OVI can serve as an alternative to traditional priors. More importantly, our analysis reveals a critical flaw in current evaluation benchmarks like T2I-CompBench++, where simply using the text embedding as a prior achieves surprisingly high scores, despite lower perceptual quality. Our constrained OVI methods improve visual fidelity over this baseline, with the Nearest-Neighbor approach proving particularly effective. It achieves quantitative scores comparable to or higher than the state-of-the-art data-efficient prior, underscoring the potential of optimization-based strategies as viable, training-free alternatives to traditional priors. The code will be publicly available upon acceptance.
Authors: Lidong Lu, Guo Chen, Zhu Wei, Yicheng Liu, Tong Lu
Abstract: We introduce Audio-Visual Affordance Grounding (AV-AG), a new task that segments object interaction regions from action sounds. Unlike existing approaches that rely on textual instructions or demonstration videos, which often limited by ambiguity or occlusion, audio provides real-time, semantically rich, and visually independent cues for affordance grounding, enabling more intuitive understanding of interaction regions. To support this task, we construct the first AV-AG dataset, comprising a large collection of action sounds, object images, and pixel-level affordance annotations. The dataset also includes an unseen subset to evaluate zero-shot generalization. Furthermore, we propose AVAGFormer, a model equipped with a semantic-conditioned cross-modal mixer and a dual-head decoder that effectively fuses audio and visual signals for mask prediction. Experiments show that AVAGFormer achieves state-of-the-art performance on AV-AG, surpassing baselines from related tasks. Comprehensive analyses highlight the distinctions between AV-AG and AVS, the benefits of end-to-end modeling, and the contribution of each component. Code and dataset have been released on https://jscslld.github.io/AVAGFormer/.
Authors: Yaokun Li, Shuaixian Wang, Mantang Guo, Jiehui Huang, Taojun Ding, Mu Hu, Kaixuan Wang, Shaojie Shen, Guang Tan
Abstract: We propose ReCamDriving, a purely vision-based, camera-controlled novel-trajectory video generation framework. While repair-based methods fail to restore complex artifacts and LiDAR-based approaches rely on sparse and incomplete cues, ReCamDriving leverages dense and scene-complete 3DGS renderings for explicit geometric guidance, achieving precise camera-controllable generation. To mitigate overfitting to restoration behaviors when conditioned on 3DGS renderings, ReCamDriving adopts a two-stage training paradigm: the first stage uses camera poses for coarse control, while the second stage incorporates 3DGS renderings for fine-grained viewpoint and geometric guidance. Furthermore, we present a 3DGS-based cross-trajectory data curation strategy to eliminate the train-test gap in camera transformation patterns, enabling scalable multi-trajectory supervision from monocular videos. Based on this strategy, we construct the ParaDrive dataset, containing over 110K parallel-trajectory video pairs. Extensive experiments demonstrate that ReCamDriving achieves state-of-the-art camera controllability and structural consistency.
Authors: Yunhong Lu, Yanhong Zeng, Haobo Li, Hao Ouyang, Qiuyu Wang, Ka Leong Cheng, Jiapeng Zhu, Hengyuan Cao, Zhipeng Zhang, Xing Zhu, Yujun Shen, Min Zhang
Abstract: Efficient streaming video generation is critical for simulating interactive and dynamic worlds. Existing methods distill few-step video diffusion models with sliding window attention, using initial frames as sink tokens to maintain attention performance and reduce error accumulation. However, video frames become overly dependent on these static tokens, resulting in copied initial frames and diminished motion dynamics. To address this, we introduce Reward Forcing, a novel framework with two key designs. First, we propose EMA-Sink, which maintains fixed-size tokens initialized from initial frames and continuously updated by fusing evicted tokens via exponential moving average as they exit the sliding window. Without additional computation cost, EMA-Sink tokens capture both long-term context and recent dynamics, preventing initial frame copying while maintaining long-horizon consistency. Second, to better distill motion dynamics from teacher models, we propose a novel Rewarded Distribution Matching Distillation (Re-DMD). Vanilla distribution matching treats every training sample equally, limiting the model's ability to prioritize dynamic content. Instead, Re-DMD biases the model's output distribution toward high-reward regions by prioritizing samples with greater dynamics rated by a vision-language model. Re-DMD significantly enhances motion quality while preserving data fidelity. We include both quantitative and qualitative experiments to show that Reward Forcing achieves state-of-the-art performance on standard benchmarks while enabling high-quality streaming video generation at 23.1 FPS on a single H100 GPU.
Authors: Priyanto Hidayatullah, Nurjannah Syakrani, Yudi Widhiyasana, Muhammad Rizqi Sholahuddin, Refdinal Tubagus, Zahri Al Adzani Hidayat, Hanri Fajar Ramadhan, Dafa Alfarizki Pratama, Farhan Muhammad Yasin
Abstract: Object detection constitutes the primary task within the domain of computer vision. It is utilized in numerous domains. Nonetheless, object detection continues to encounter the issue of catastrophic forgetting. The model must be retrained whenever new products are introduced, utilizing not only the new products dataset but also the entirety of the previous dataset. The outcome is obvious: increasing model training expenses and significant time consumption. In numerous sectors, particularly retail checkout, the frequent introduction of new products presents a great challenge. This study introduces Zero-Retraining Based Recognition and Object Detection (ZeBROD), a methodology designed to address the issue of catastrophic forgetting by integrating YOLO11n for object localization with DeIT and Proxy Anchor Loss for feature extraction and metric learning. For classification, we utilize cosine similarity between the embedding features of the target product and those in the Qdrant vector database. In a case study conducted in a retail store with 140 products, the experimental results demonstrate that our proposed framework achieves encouraging accuracy, whether for detecting new or existing products. Furthermore, without retraining, the training duration difference is significant. We achieve almost 3 times the training time efficiency compared to classical object detection approaches. This efficiency escalates as additional new products are added to the product database. The average inference time is 580 ms per image containing multiple products, on an edge device, validating the proposed framework's feasibility for practical use.
Authors: Shresth Grover, Priyank Pathak, Akash Kumar, Vibhav Vineet, Yogesh S Rawat
Abstract: Large-scale Vision-Language Models (VLMs) exhibit impressive complex reasoning capabilities but remain largely unexplored in visual sequential planning, i.e., executing multi-step actions towards a goal. Additionally, practical sequential planning often involves non-optimal (erroneous) steps, challenging VLMs to detect and correct such steps. We propose Corrective Sequential Planning Benchmark (CoSPlan) to evaluate VLMs in error-prone, vision-based sequential planning tasks across 4 domains: maze navigation, block rearrangement, image reconstruction,and object reorganization. CoSPlan assesses two key abilities: Error Detection (identifying non-optimal action) and Step Completion (correcting and completing action sequences to reach the goal). Despite using state-of-the-art reasoning techniques such as Chain-of-Thought and Scene Graphs, VLMs (e.g. Intern-VLM and Qwen2) struggle on CoSPlan, failing to leverage contextual cues to reach goals. Addressing this, we propose a novel training-free method, Scene Graph Incremental updates (SGI), which introduces intermediate reasoning steps between the initial and goal states. SGI helps VLMs reason about sequences, yielding an average performance gain of 5.2%. In addition to enhancing reliability in corrective sequential planning, SGI generalizes to traditional planning tasks such as Plan-Bench and VQA. Project Page : https://shroglck.github.io/cos_plan/
Authors: Zhankuo Xu, Chaoran Feng, Yingtao Li, Jianbin Zhao, Jiashu Yang, Wangbo Yu, Li Yuan, Yonghong Tian
Abstract: 3D Gaussian Splatting (3DGS) has emerged as a state-of-the-art method for novel view synthesis. However, its performance heavily relies on dense, high-quality input imagery, an assumption that is often violated in real-world applications, where data is typically sparse and motion-blurred. These two issues create a vicious cycle: sparse views ignore the multi-view constraints necessary to resolve motion blur, while motion blur erases high-frequency details crucial for aligning the limited views. Thus, reconstruction often fails catastrophically, with fragmented views and a low-frequency bias. To break this cycle, we introduce CoherentGS, a novel framework for high-fidelity 3D reconstruction from sparse and blurry images. Our key insight is to address these compound degradations using a dual-prior strategy. Specifically, we combine two pre-trained generative models: a specialized deblurring network for restoring sharp details and providing photometric guidance, and a diffusion model that offers geometric priors to fill in unobserved regions of the scene. This dual-prior strategy is supported by several key techniques, including a consistency-guided camera exploration module that adaptively guides the generative process, and a depth regularization loss that ensures geometric plausibility. We evaluate CoherentGS through both quantitative and qualitative experiments on synthetic and real-world scenes, using as few as 3, 6, and 9 input views. Our results demonstrate that CoherentGS significantly outperforms existing methods, setting a new state-of-the-art for this challenging task. The code and video demos are available at https://potatobigroom.github.io/CoherentGS/.
Authors: Wen-Jue He, Xiaofeng Zhu, Zheng Zhang
Abstract: Incomplete multi-modal emotion recognition (IMER) aims at understanding human intentions and sentiments by comprehensively exploring the partially observed multi-source data. Although the multi-modal data is expected to provide more abundant information, the performance gap and modality under-optimization problem hinder effective multi-modal learning in practice, and are exacerbated in the confrontation of the missing data. To address this issue, we devise a novel Cross-modal Prompting (ComP) method, which emphasizes coherent information by enhancing modality-specific features and improves the overall recognition accuracy by boosting each modality's performance. Specifically, a progressive prompt generation module with a dynamic gradient modulator is proposed to produce concise and consistent modality semantic cues. Meanwhile, cross-modal knowledge propagation selectively amplifies the consistent information in modality features with the delivered prompts to enhance the discrimination of the modality-specific output. Additionally, a coordinator is designed to dynamically re-weight the modality outputs as a complement to the balance strategy to improve the model's efficacy. Extensive experiments on 4 datasets with 7 SOTA methods under different missing rates validate the effectiveness of our proposed method.
Authors: Ahmed Abul Hasanaath, Hamzah Luqman
Abstract: Continuous sign language recognition (CSLR) requires precise spatio-temporal modeling to accurately recognize sequences of gestures in videos. Existing frameworks often rely on CNN-based spatial backbones combined with temporal convolution or recurrent modules. These techniques fail in capturing fine-grained hand and facial cues and modeling long-range temporal dependencies. To address these limitations, we propose the Unified Spatio-Temporal Modeling (USTM) framework, a spatio-temporal encoder that effectively models complex patterns using a combination of a Swin Transformer backbone enhanced with lightweight temporal adapter with positional embeddings (TAPE). Our framework captures fine-grained spatial features alongside short and long-term temporal context, enabling robust sign language recognition from RGB videos without relying on multi-stream inputs or auxiliary modalities. Extensive experiments on benchmarked datasets including PHOENIX14, PHOENIX14T, and CSL-Daily demonstrate that USTM achieves state-of-the-art performance against RGB-based as well as multi-modal CSLR approaches, while maintaining competitive performance against multi-stream approaches. These results highlight the strength and efficacy of the USTM framework for CSLR. The code is available at https://github.com/gufranSabri/USTM
Authors: Zhenguo Zhang, Haohan Zheng, Yishen Wang, Le Xu, Tianchen Deng, Xuefeng Chen, Qu Chen, Bo Zhang, Wuxiong Huang
Abstract: The deployment of Vision-Language Models (VLMs) in safety-critical domains like autonomous driving (AD) is critically hindered by reliability failures, most notably object hallucination. This failure stems from their reliance on ungrounded, text-based Chain-of-Thought (CoT) reasoning. While existing multi-modal CoT approaches attempt mitigation, they suffer from two fundamental flaws: (1) decoupled perception and reasoning stages that prevent end-to-end joint optimization, and (2) reliance on expensive, dense localization labels. Thus we introduce OmniDrive-R1, an end-to-end VLM framework designed for autonomous driving, which unifies perception and reasoning through an interleaved Multi-modal Chain-of-Thought (iMCoT) mechanism. Our core innovation is an Reinforcement-driven visual grounding capability, enabling the model to autonomously direct its attention and "zoom in" on critical regions for fine-grained analysis. This capability is enabled by our pure two-stage reinforcement learning training pipeline and Clip-GRPO algorithm. Crucially, Clip-GRPO introduces an annotation-free, process-based grounding reward. This reward not only eliminates the need for dense labels but also circumvents the instability of external tool calls by enforcing real-time cross-modal consistency between the visual focus and the textual reasoning. Extensive experiments on DriveLMM-o1 demonstrate our model's significant improvements. Compared to the baseline Qwen2.5VL-7B, OmniDrive-R1 improves the overall reasoning score from 51.77% to 80.35%, and the final answer accuracy from 37.81% to 73.62%.
Authors: Bozhou Li, Sihan Yang, Yushuo Guan, Ruichuan An, Xinlong Chen, Yang Shi, Pengfei Wan, Wentao Zhang, Yuanxing zhang
Abstract: The text encoder is a critical component of text-to-image and text-to-video diffusion models, fundamentally determining the semantic fidelity of the generated content. However, its development has been hindered by two major challenges: the lack of an efficient evaluation framework that reliably predicts downstream generation performance, and the difficulty of effectively adapting pretrained language models for visual synthesis. To address these issues, we introduce GRAN-TED, a paradigm to Generate Robust, Aligned, and Nuanced Text Embeddings for Diffusion models. Our contribution is twofold. First, we propose TED-6K, a novel text-only benchmark that enables efficient and robust assessment of an encoder's representational quality without requiring costly end-to-end model training. We demonstrate that performance on TED-6K, standardized via a lightweight, unified adapter, strongly correlates with an encoder's effectiveness in downstream generation tasks. Notably, under our experimental setup, compared with training a diffusion model from scratch, evaluating with TED-6K is about \textbf{750$\times$ faster}. Second, guided by this validated framework, we develop a superior text encoder using a novel two-stage training paradigm. This process involves an initial fine-tuning stage on a Multimodal Large Language Model for better visual representation, followed by a layer-wise weighting method to extract more nuanced and potent text features. Our experiments show that the resulting GRAN-TED encoder not only achieves state-of-the-art performance on TED-6K but also leads to demonstrable performance gains in text-to-image and text-to-video generation. Our TED-6K dataset and evaluation code are available at the following link: https://anonymous.4open.science/r/GRAN-TED-4FCC/.
Authors: Shangxun Li, Youngjung Uh
Abstract: Text-to-image diffusion models excel at generating high-quality images from natural language descriptions but often fail to preserve subject consistency across multiple outputs, limiting their use in visual storytelling. Existing approaches rely on model fine-tuning or image conditioning, which are computationally expensive and require per-subject optimization. 1Prompt1Story, a training-free approach, concatenates all scene descriptions into a single prompt and rescales token embeddings, but it suffers from semantic leakage, where embeddings across frames become entangled, causing text misalignment. In this paper, we propose a simple yet effective training-free approach that addresses semantic entanglement from a geometric perspective by refining text embeddings to suppress unwanted semantics. Extensive experiments prove that our approach significantly improves both subject consistency and text alignment over existing baselines.
Authors: Sarah Rastegar, Violeta Chatalbasheva, Sieger Falkena, Anuj Singh, Yanbo Wang, Tejas Gokhale, Hamid Palangi, Hadi Jamali-Rad
Abstract: Text-to-image (T2I) diffusion models generate high-quality images but often fail to capture the spatial relations specified in text prompts. This limitation can be traced to two factors: lack of fine-grained spatial supervision in training data and inability of text embeddings to encode spatial semantics. We introduce InfSplign, a training-free inference-time method that improves spatial alignment by adjusting the noise through a compound loss in every denoising step. Proposed loss leverages different levels of cross-attention maps extracted from the backbone decoder to enforce accurate object placement and a balanced object presence during sampling. The method is lightweight, plug-and-play, and compatible with any diffusion backbone. Our comprehensive evaluations on VISOR and T2I-CompBench show that InfSplign establishes a new state-of-the-art (to the best of our knowledge), achieving substantial performance gains over the strongest existing inference-time baselines and even outperforming the fine-tuning-based methods. Codebase is available at GitHub.
Authors: Kaixing Yang, Jiashu Zhu, Xulong Tang, Ziqiao Peng, Xiangyue Zhang, Puwei Wang, Jiahong Wu, Xiangxiang Chu, Hongyan Liu, Jun He
Abstract: With the rise of online dance-video platforms and rapid advances in AI-generated content (AIGC), music-driven dance generation has emerged as a compelling research direction. Despite substantial progress in related domains such as music-driven 3D dance generation, pose-driven image animation, and audio-driven talking-head synthesis, existing methods cannot be directly adapted to this task. Moreover, the limited studies in this area still struggle to jointly achieve high-quality visual appearance and realistic human motion. Accordingly, we present MACE-Dance, a music-driven dance video generation framework with cascaded Mixture-of-Experts (MoE). The Motion Expert performs music-to-3D motion generation while enforcing kinematic plausibility and artistic expressiveness, whereas the Appearance Expert carries out motion- and reference-conditioned video synthesis, preserving visual identity with spatiotemporal coherence. Specifically, the Motion Expert adopts a diffusion model with a BiMamba-Transformer hybrid architecture and a Guidance-Free Training (GFT) strategy, achieving state-of-the-art (SOTA) performance in 3D dance generation. The Appearance Expert employs a decoupled kinematic-aesthetic fine-tuning strategy, achieving state-of-the-art (SOTA) performance in pose-driven image animation. To better benchmark this task, we curate a large-scale and diverse dataset and design a motion-appearance evaluation protocol. Based on this protocol, MACE-Dance also achieves state-of-the-art performance. Project page: https://macedance.github.io/
Authors: Meng Chu, Senqiao Yang, Haoxuan Che, Suiyun Zhang, Xichen Zhang, Shaozuo Yu, Haokun Gui, Zhefan Rao, Dandan Tu, Rui Liu, Jiaya Jia
Abstract: Generative models can now produce photorealistic imagery, yet they still struggle with the long, multi-goal prompts that professional designers issue. To expose this gap and better evaluate models' performance in real-world settings, we introduce Long Goal Bench (LGBench), a 2,000-task suite (1,000 T2I and 1,000 I2I) whose average instruction contains 18 to 22 tightly coupled goals spanning global layout, local object placement, typography, and logo fidelity. We find that even state-of-the-art models satisfy fewer than 72 percent of the goals and routinely miss localized edits, confirming the brittleness of current pipelines. To address this, we present VisionDirector, a training-free vision-language supervisor that (i) extracts structured goals from long instructions, (ii) dynamically decides between one-shot generation and staged edits, (iii) runs micro-grid sampling with semantic verification and rollback after every edit, and (iv) logs goal-level rewards. We further fine-tune the planner with Group Relative Policy Optimization, yielding shorter edit trajectories (3.1 versus 4.2 steps) and stronger alignment. VisionDirector achieves new state of the art on GenEval (plus 7 percent overall) and ImgEdit (plus 0.07 absolute) while producing consistent qualitative improvements on typography, multi-object scenes, and pose editing.
Authors: Mingrui Wu, Zhaozhi Wang, Fangjinhua Wang, Jiaolong Yang, Marc Pollefeys, Tong Zhang
Abstract: While Multimodal Large Language Models (MLLMs) have achieved impressive performance on semantic tasks, their spatial intelligence--crucial for robust and grounded AI systems--remains underdeveloped. Existing benchmarks fall short of diagnosing this limitation: they either focus on overly simplified qualitative reasoning or rely on domain-specific indoor data, constrained by the lack of outdoor datasets with verifiable metric ground truth. To bridge this gap, we introduce a large-scale benchmark built from pedestrian-perspective videos captured with synchronized stereo cameras, LiDAR, and IMU/GPS sensors. This dataset provides metrically precise 3D information, enabling the automatic generation of spatial reasoning questions that span a hierarchical spectrum--from qualitative relational reasoning to quantitative metric and kinematic understanding. Evaluations reveal that the performance gains observed in structured indoor benchmarks vanish in open-world settings. Further analysis using synthetic abnormal scenes and blinding tests confirms that current MLLMs depend heavily on linguistic priors instead of grounded visual reasoning. Our benchmark thus provides a principled platform for diagnosing these limitations and advancing physically grounded spatial intelligence.
Authors: SaiKiran Tedla, Zhoutong Zhang, Xuaner Zhang, Shumian Xin
Abstract: Focus is a cornerstone of photography, yet autofocus systems often fail to capture the intended subject, and users frequently wish to adjust focus after capture. We introduce a novel method for realistic post-capture refocusing using video diffusion models. From a single defocused image, our approach generates a perceptually accurate focal stack, represented as a video sequence, enabling interactive refocusing and unlocking a range of downstream applications. We release a large-scale focal stack dataset acquired under diverse real-world smartphone conditions to support this work and future research. Our method consistently outperforms existing approaches in both perceptual quality and robustness across challenging scenarios, paving the way for more advanced focus-editing capabilities in everyday photography. Code and data are available at www.learn2refocus.github.io
Authors: Chang Sun, Dongliang Xie, Wanpeng Xie, Bo Qin, Hong Yang
Abstract: Visual speech recognition (VSR) aims to transcribe spoken content from silent lip-motion videos and is particularly challenging in Mandarin due to severe viseme ambiguity and pervasive homophones. We propose VALLR-Pin, a two-stage Mandarin VSR framework that extends the VALLR architecture by explicitly incorporating Pinyin as an intermediate representation. In the first stage, a shared visual encoder feeds dual decoders that jointly predict Mandarin characters and their corresponding Pinyin sequences, encouraging more robust visual-linguistic representations. In the second stage, an LLM-based refinement module takes the predicted Pinyin sequence together with an N-best list of character hypotheses to resolve homophone-induced ambiguities. To further adapt the LLM to visual recognition errors, we fine-tune it on synthetic instruction data constructed from model-generated Pinyin-text pairs, enabling error-aware correction. Experiments on public Mandarin VSR benchmarks demonstrate that VALLR-Pin consistently improves transcription accuracy under multi-speaker conditions, highlighting the effectiveness of combining phonetic guidance with lightweight LLM refinement.
Authors: Alireza Moayedikia, Sattar Dorafshan
Abstract: Deteriorating civil infrastructure requires automated inspection techniques overcoming limitations of visual assessment. While Ground Penetrating Radar and Infrared Thermography enable subsurface defect detection, single modal approaches face complementary constraints radar struggles with moisture and shallow defects, while thermography exhibits weather dependency and limited depth. This paper presents a multi modal attention network fusing radar temporal patterns with thermal spatial signatures for bridge deck delamination detection. Our architecture introduces temporal attention for radar processing, spatial attention for thermal features, and cross modal fusion with learnable embeddings discovering complementary defect patterns invisible to individual sensors. We incorporate uncertainty quantification through Monte Carlo dropout and learned variance estimation, decomposing uncertainty into epistemic and aleatoric components for safety critical decisions. Experiments on five bridge datasets reveal that on balanced to moderately imbalanced data, our approach substantially outperforms baselines in accuracy and AUC representing meaningful improvements over single modal and concatenation based fusion. Ablation studies demonstrate cross modal attention provides critical gains beyond within modality attention, while multi head mechanisms achieve improved calibration. Uncertainty quantification reduces calibration error, enabling selective prediction by rejecting uncertain cases. However, under extreme class imbalance, attention mechanisms show vulnerability to majority class collapse. These findings provide actionable guidance: attention based architecture performs well across typical scenarios, while extreme imbalance requires specialized techniques. Our system maintains deployment efficiency, enabling real time inspection with characterized capabilities and limitations.
Authors: Xiangzuo Wu, Chengwei Ren, Jun Zhou, Xiu Li, Yuan Liu
Abstract: Multi-view inverse rendering aims to recover geometry, materials, and illumination consistently across multiple viewpoints. When applied to multi-view images, existing single-view approaches often ignore cross-view relationships, leading to inconsistent results. In contrast, multi-view optimization methods rely on slow differentiable rendering and per-scene refinement, making them computationally expensive and hard to scale. To address these limitations, we introduce a feed-forward multi-view inverse rendering framework that directly predicts spatially varying albedo, metallic, roughness, diffuse shading, and surface normals from sequences of RGB images. By alternating attention across views, our model captures both intra-view long-range lighting interactions and inter-view material consistency, enabling coherent scene-level reasoning within a single forward pass. Due to the scarcity of real-world training data, models trained on existing synthetic datasets often struggle to generalize to real-world scenes. To overcome this limitation, we propose a consistency-based finetuning strategy that leverages unlabeled real-world videos to enhance both multi-view coherence and robustness under in-the-wild conditions. Extensive experiments on benchmark datasets demonstrate that our method achieves state-of-the-art performance in terms of multi-view consistency, material and normal estimation quality, and generalization to real-world imagery. Project page: https://maddog241.github.io/mvinverse-page/
Authors: Rui-qing Sun, Xingshan Yao, Tian Lan, Jia-Ling Shi, Chen-Hao Cui, Hui-Yang Zhao, Zhijing Wu, Chen Yang, Xian-Ling Mao
Abstract: State-of-the-art 3D-field video-referenced Talking Face Generation (TFG) methods synthesize high-fidelity personalized talking-face videos in real time by modeling 3D geometry and appearance from reference portrait video. This capability raises significant privacy concerns regarding malicious misuse of personal portraits. However, no efficient defense framework exists to protect such videos against 3D-field TFG methods. While image-based defenses could apply per-frame 2D perturbations, they incur prohibitive computational costs, severe video quality degradation, failing to disrupt 3D information for video protection. To address this, we propose a novel and efficient video defense framework against 3D-field TFG methods, which protects portrait video by perturbing the 3D information acquisition process while maintain high-fidelity video quality. Specifically, our method introduces: (1) a similarity-guided parameter sharing mechanism for computational efficiency, and (2) a multi-scale dual-domain attention module to jointly optimize spatial-frequency perturbations. Extensive experiments demonstrate that our proposed framework exhibits strong defense capability and achieves a 47x acceleration over the fastest baseline while maintaining high fidelity. Moreover, it remains robust against scaling operations and state-of-the-art purification attacks, and the effectiveness of our design choices is further validated through ablation studies. Our project is available at https://github.com/Richen7418/VDF.
Authors: Tianchen Deng, Xun Chen, Ziming Li, Hongming Shen, Danwei Wang, Javier Civera, Hesheng Wang
Abstract: Visual Place Recognition (VPR) has been traditionally formulated as a single-image retrieval task. Using multiple views offers clear advantages, yet this setting remains relatively underexplored and existing methods often struggle to generalize across diverse environments. In this work we introduce UniPR-3D, the first VPR architecture that effectively integrates information from multiple views. UniPR-3D builds on a VGGT backbone capable of encoding multi-view 3D representations, which we adapt by designing feature aggregators and fine-tune for the place recognition task. To construct our descriptor, we jointly leverage the 3D tokens and intermediate 2D tokens produced by VGGT. Based on their distinct characteristics, we design dedicated aggregation modules for 2D and 3D features, allowing our descriptor to capture fine-grained texture cues while also reasoning across viewpoints. To further enhance generalization, we incorporate both single- and multi-frame aggregation schemes, along with a variable-length sequence retrieval strategy. Our experiments show that UniPR-3D sets a new state of the art, outperforming both single- and multi-view baselines and highlighting the effectiveness of geometry-grounded tokens for VPR. Our code and models will be made publicly available on Github https://github.com/dtc111111/UniPR-3D.
Authors: Zhiwang Zhou, Yuandong Pu, Xuming He, Yidi Liu, Yixin Chen, Junchao Gong, Xiang Zhuang, Wanghan Xu, Qinglong Cao, Shixiang Tang, Yihao Liu, Wenlong Zhang, Lei Bai
Abstract: Weather modeling requires both accurate prediction and mechanistic interpretation, yet existing methods treat these goals in isolation, separating generation from understanding. To address this gap, we present Omni-Weather, the first multimodal foundation model that unifies weather generation and understanding within a single architecture. Omni-Weather integrates a radar encoder for weather generation tasks, followed by unified processing using a shared self-attention mechanism. Moreover, we construct a Chain-of-Thought dataset for causal reasoning in weather generation, enabling interpretable outputs and improved perceptual quality. Extensive experiments show Omni-Weather achieves state-of-the-art performance in both weather generation and understanding. Our findings further indicate that generative and understanding tasks in the weather domain can mutually enhance each other. Omni-Weather also demonstrates the feasibility and value of unifying weather generation and understanding.
Authors: Zhan Chen, Zile Guo, Enze Zhu, Peirong Zhang, Xiaoxuan Liu, Lei Wang, Yidan Zhang
Abstract: Video prediction is plagued by a fundamental trilemma: achieving high-resolution and perceptual quality typically comes at the cost of real-time speed, hindering its use in latency-critical applications. This challenge is most acute for autonomous UAVs in dense urban environments, where foreseeing events from high-resolution imagery is non-negotiable for safety. Existing methods, reliant on iterative generation (diffusion, autoregressive models) or quadratic-complexity attention, fail to meet these stringent demands on edge hardware. To break this long-standing trade-off, we introduce RAPTOR, a video prediction architecture that achieves real-time, high-resolution performance. RAPTOR's single-pass design avoids the error accumulation and latency of iterative approaches. Its core innovation is Efficient Video Attention (EVA), a novel translator module that factorizes spatiotemporal modeling. Instead of processing flattened spacetime tokens with $O((ST)^2)$ or $O(ST)$ complexity, EVA alternates operations along the spatial (S) and temporal (T) axes. This factorization reduces the time complexity to $O(S + T)$ and memory complexity to $O(max(S, T))$, enabling global context modeling at $512^2$ resolution and beyond, operating directly on dense feature maps with a patch-free design. Complementing this architecture is a 3-stage training curriculum that progressively refines predictions from coarse structure to sharp, temporally coherent details. Experiments show RAPTOR is the first predictor to exceed 30 FPS on a Jetson AGX Orin for $512^2$ video, setting a new state-of-the-art on UAVid, KTH, and a custom high-resolution dataset in PSNR, SSIM, and LPIPS. Critically, RAPTOR boosts the mission success rate in a real-world UAV navigation task by 18%, paving the way for safer and more anticipatory embodied agents.
Authors: Steven Xiao, Xindi Zhang, Dechao Meng, Qi Wang, Peng Zhang, Bang Zhang
Abstract: Real-time portrait animation is essential for interactive applications such as virtual assistants and live avatars, requiring high visual fidelity, temporal coherence, ultra-low latency, and responsive control from dynamic inputs like reference images and driving signals. While diffusion-based models achieve strong quality, their non-causal nature hinders streaming deployment. Causal autoregressive video generation approaches enable efficient frame-by-frame generation but suffer from error accumulation, motion discontinuities at chunk boundaries, and degraded long-term consistency. In this work, we present a novel streaming framework named Knot Forcing for real-time portrait animation that addresses these challenges through three key designs: (1) a chunk-wise generation strategy with global identity preservation via cached KV states of the reference image and local temporal modeling using sliding window attention; (2) a temporal knot module that overlaps adjacent chunks and propagates spatio-temporal cues via image-to-video conditioning to smooth inter-chunk motion transitions; and (3) A "running ahead" mechanism that dynamically updates the reference frame's temporal coordinate during inference, keeping its semantic context ahead of the current rollout frame to support long-term coherence. Knot Forcing enables high-fidelity, temporally consistent, and interactive portrait animation over infinite sequences, achieving real-time performance with strong visual stability on consumer-grade GPUs.
Authors: Yunxin Li, Zhenyu Liu, Baotian Hu, Wei Wang, Yuxin Ding, Xiaochun Cao, Min Zhang
Abstract: Recent advancements in multimodal large language models (MLLMs) have achieved significant multimodal generation capabilities, akin to GPT-4. These models predominantly map visual information into language representation space, leveraging the vast knowledge and powerful text generation abilities of LLMs to produce multimodal instruction-following responses. We could term this method as LLMs for Vision because of its employing LLMs for visual understanding and reasoning, yet observe that these MLLMs neglect the potential of harnessing visual knowledge to enhance the overall capabilities of LLMs, which could be regarded as Vision Enhancing LLMs. In this paper, we propose an approach called MKS2, aimed at enhancing LLMs through empowering Multimodal Knowledge Storage and Sharing in LLMs. Specifically, we introduce Modular Visual Memory (MVM), a component integrated into the internal blocks of LLMs, designed to store open-world visual information efficiently. Additionally, we present a soft Mixture of Multimodal Experts (MoMEs) architecture in LLMs to invoke multimodal knowledge collaboration during text generation. Our comprehensive experiments demonstrate that MKS2 substantially augments the reasoning capabilities of LLMs in contexts necessitating physical or commonsense knowledge. It also delivers competitive results on image-text understanding multimodal benchmarks. The codes will be available at: https://github.com/HITsz-TMG/MKS2-Multimodal-Knowledge-Storage-and-Sharing
URLs: https://github.com/HITsz-TMG/MKS2-Multimodal-Knowledge-Storage-and-Sharing
Authors: Hao Zhang, Qi Wang, Jian Sun, Zhijie Wen, Jun Shi, Shihui Ying
Abstract: Magnetic Resonance Imaging (MRI) is widely used in clinical practice, but suffered from prolonged acquisition time. Although deep learning methods have been proposed to accelerate acquisition and demonstrate promising performance, they rely on high-quality fully-sampled datasets for training in a supervised manner. However, such datasets are time-consuming and expensive-to-collect, which constrains their broader applications. On the other hand, self-supervised methods offer an alternative by enabling learning from under-sampled data alone, but most existing methods rely on further partitioned under-sampled k-space data as model's input for training, resulting in a loss of valuable information. Additionally, their models have not fully incorporated image priors, leading to degraded reconstruction performance. In this paper, we propose a novel re-visible dual-domain self-supervised deep unfolding network to address these issues when only under-sampled datasets are available. Specifically, by incorporating re-visible dual-domain loss, all under-sampled k-space data are utilized during training to mitigate information loss caused by further partitioning. This design enables the model to implicitly adapt to all under-sampled k-space data as input. Additionally, we design a deep unfolding network based on Chambolle and Pock Proximal Point Algorithm (DUN-CP-PPA) to achieve end-to-end reconstruction, incorporating imaging physics and image priors to guide the reconstruction process. By employing a Spatial-Frequency Feature Extraction (SFFE) block to capture global and local feature representation, we enhance the model's efficiency to learn comprehensive image priors. Experiments conducted on the fastMRI and IXI datasets demonstrate that our method significantly outperforms state-of-the-art approaches in terms of reconstruction performance.
Authors: Silin Chen, Andy Liu, Kangjian Di, Yichu Xu, Han-Jia Ye, Wenhan Luo, Ningmu Zou
Abstract: Unsupervised image anomaly detection (UAD) has become a critical process in industrial and medical applications, but it faces growing challenges due to increasing concerns over data privacy. The limited class diversity inherent to one-class classification tasks, combined with distribution biases caused by variations in products across and within clients, poses significant challenges for preserving data privacy with federated UAD. Thus, this article proposes an efficient federated learning method with dynamic memory and memory-reduce for unsupervised image anomaly detection, called FedDyMem. Considering all client data belongs to a single class (i.e., normal sample) in UAD and the distribution of intra-class features demonstrates significant skewness, FedDyMem facilitates knowledge sharing between the client and server through the client's dynamic memory bank instead of model parameters. In the local clients, a memory generator and a metric loss are employed to improve the consistency of the feature distribution for normal samples, leveraging the local model to update the memory bank dynamically. For efficient communication and data privacy, a memory-reduce method based on weighted averages is proposed to significantly decrease the scale of memory banks. This reduced representation inherently, thereby mitigating the risk of data reconstruction. On the server, global memory is constructed and distributed to individual clients through k-means aggregation. Experiments conducted on six industrial and medical datasets, comprising a mixture of six products or health screening types derived from eleven public datasets, demonstrate the effectiveness of FedDyMem.
Authors: Luohong Wu, Matthias Seibold, Nicola A. Cavalcanti, Giuseppe Loggia, Lisa Reissner, Bastian Sigrist, Jonas Hein, Lilian Calvet, Arnd Vieh\"ofer, Philipp F\"urnstahl
Abstract: Bone surface reconstruction is an essential component of computer-assisted orthopedic surgery(CAOS), forming the foundation for both preoperative planning and intraoperative guidance. Compared to traditional imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI),ultrasound, an emerging CAOS technology, provides a radiation-free, cost-effective, and portable alternative. While ultrasound offers new opportunities in CAOS, technical shortcomings continue to hinder its translation into surgery. In particular, due to the inherent limitations of ultrasound imaging, B-mode ultrasound typically captures only partial bone surfaces. The inter- and intra-operator variability in ultrasound scanning further increases the complexity of the data. Existing reconstruction methods struggle with such challenging data, leading to increased reconstruction errors and artifacts, such as holes and inflated structures. Effective techniques for accurately reconstructing open bone surfaces from real-world 3D ultrasound volumes remain lacking. We propose UltraBoneUDF, a self-supervised framework specifically designed for reconstructing open bone surfaces from ultrasound data. It learns unsigned distance functions (UDFs) from 3D ultrasound data. In addition, we present a novel loss function based on local tangent plane optimization that substantially improves surface reconstruction quality. UltraBoneUDF and competing models are benchmarked on three open-source datasets and further evaluated through ablation studies. Qualitative results demonstrate the limitations of the state-of-the-art methods. Quantitatively, UltraBoneUDF achieves comparable or lower bi-directional Chamfer distance across three datasets with fewer parameters: 1.60 mm on the UltraBones100k dataset (~25.5% improvement), 0.21 mm on the OpenBoneCT dataset, and 0.18 mm on the ClosedBoneCT dataset.
Authors: Chenyu Su, Weiwei Shang, Chen Qian, Fei Zhang, Shuang Cong
Abstract: Fine-grained robotic manipulation requires grounding natural language into appropriate affordance targets. However, most existing methods driven by foundation models often compress rich semantics into oversimplified affordances, preventing exploitation of implicit semantic information. To address these challenges, we present ReSemAct, a novel unified manipulation framework that introduces Semantic Structuring and Affordance Refinement (SSAR), powered by the automated synergistic reasoning between Multimodal Large Language Models (MLLMs) and Vision Foundation Models (VFMs). Specifically, the Semantic Structuring module derives a unified semantic affordance description from natural language and RGB observations, organizing affordance regions, implicit functional intent, and coarse affordance anchors into a structured representation for downstream refinement. Building upon this specification, the Affordance Refinement strategy instantiates two complementary flows that separately specialize geometry and position, yielding fine-grained affordance targets. These refined targets are then encoded as real-time joint-space optimization objectives, enabling reactive and robust manipulation in dynamic environments. Extensive simulation and real-world experiments are conducted in semantically rich household and sparse chemical lab environments. The results demonstrate that ReSemAct performs diverse tasks under zero-shot conditions, showcasing the robustness of SSAR with foundation models in fine-grained manipulation. Code and videos at https://github.com/scy-v/ReSemAct and https://resemact.github.io.
URLs: https://github.com/scy-v/ReSemAct, https://resemact.github.io.
Authors: Raman Jha, Yang Zhou, Giuseppe Loianno
Abstract: In this paper, we propose an adaptive keyframe selection method for improved 3D scene reconstruction in dynamic environments. The proposed method integrates two complementary modules: an error-based selection module utilizing photometric and structural similarity (SSIM) errors, and a momentum-based update module that dynamically adjusts keyframe selection thresholds according to scene motion dynamics. By dynamically curating the most informative frames, our approach addresses a key data bottleneck in real-time perception. This allows for the creation of high-quality 3D world representations from a compressed data stream, a critical step towards scalable robot learning and deployment in complex, dynamic environments. Experimental results demonstrate significant improvements over traditional static keyframe selection strategies, such as fixed temporal intervals or uniform frame skipping. These findings highlight a meaningful advancement toward adaptive perception systems that can dynamically respond to complex and evolving visual scenes. We evaluate our proposed adaptive keyframe selection module on two recent state-of-the-art 3D reconstruction networks, Spann3r and CUT3R, and observe consistent improvements in reconstruction quality across both frameworks. Furthermore, an extensive ablation study confirms the effectiveness of each individual component in our method, underlining their contribution to the overall performance gains.
Authors: Yujia Fu, Zhiyu Dong, Tianwen Qian, Chenye Zheng, Danian Ji, Linhai Zhuo
Abstract: Accurate assessment of bowel cleanliness is essential for effective colonoscopy procedures. The Boston Bowel Preparation Scale (BBPS) offers a standardized scoring system but suffers from subjectivity and inter-observer variability when performed manually. In this paper, to support robust training and evaluation, we construct a high-quality colonoscopy dataset comprising 2,240 images from 517 subjects, annotated with expert-agreed BBPS scores. We propose a novel automated BBPS scoring framework that leverages the CLIP model with adapter-based transfer learning and a dedicated fecal-feature extraction branch. Our method fuses global visual features with stool-related textual priors to improve the accuracy of bowel cleanliness evaluation without requiring explicit segmentation. Extensive experiments on both our dataset and the public NERTHU dataset demonstrate the superiority of our approach over existing baselines, highlighting its potential for clinical deployment in computer-aided colonoscopy analysis.