Authors: Bhavin Jawade, Joao V. B. Soares, Kapil Thadani, Deen Dayal Mohan, Amir Erfan Eshratifar, Benjamin Culpepper, Paloma de Juan, Srirangaraj Setlur, Venu Govindaraju
Abstract: Compositional image retrieval (CIR) is a multimodal learning task where a model combines a query image with a user-provided text modification to retrieve a target image. CIR finds applications in a variety of domains including product retrieval (e-commerce) and web search. Existing methods primarily focus on fully-supervised learning, wherein models are trained on datasets of labeled triplets such as FashionIQ and CIRR. This poses two significant challenges: (i) curating such triplet datasets is labor intensive; and (ii) models lack generalization to unseen objects and domains. In this work, we propose SCOT (Self-supervised COmpositional Training), a novel zero-shot compositional pretraining strategy that combines existing large image-text pair datasets with the generative capabilities of large language models to contrastively train an embedding composition network. Specifically, we show that the text embedding from a large-scale contrastively-pretrained vision-language model can be utilized as proxy target supervision during compositional pretraining, replacing the target image embedding. In zero-shot settings, this strategy surpasses SOTA zero-shot compositional retrieval methods as well as many fully-supervised methods on standard benchmarks such as FashionIQ and CIRR.
Authors: Shijian Xu
Abstract: Empirical risk minimization (ERM) is not robust to changes in the distribution of data. When the distribution of test data is different from that of training data, the problem is known as out-of-distribution generalization. Recently, two techniques have been developed for addressing out-of-distribution generalization in computer vision: weight averaging (WA) and sharpness-aware minimization (SAM). WA involves training multiple models with different hyperparameters and then averaging the weights of these models, which can significantly improve out-of-distribution generalization performance. SAM optimizes a neural network to find minima in flat regions, which have been proven to perform well under distribution shifts. While these techniques have made great progress, there is still room for improvement and further exploration. In this thesis, we propose increasing the model diversity in WA explicitly by introducing gradient similarity as a loss regularizer to further improve out-of-distribution generalization performance. We also propose combining WA and SAM to solve the problem of few-shot domain adaptation. Our extensive experiments on digits datasets (MNIST, SVHN, USPS, MNIST-M) and other domain adaptation datasets (VLCS, PACS) show that combining WA and SAM leads to improved out-of-distribution generalization performance and significantly increases few-shot domain adaptation accuracy.
Authors: Hansoo Park, Chanwoo Kim, Jihyeon Kim, Hoseong Cho, Nhat Nguyen Bao Truong, Taehwan Kim, Seungryul Baek
Abstract: RGB-based 3D pose estimation methods have been successful with the development of deep learning and the emergence of high-quality 3D pose datasets. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. This problem might be alleviated by involving diverse data during training, however it is non-trivial to collect such diverse data with corresponding labels (i.e. 3D pose). In this paper, we introduced an unsupervised domain adaptation framework for 3D pose estimation that utilizes the unlabeled data in addition to labeled data via masked image modeling (MIM) framework. Foreground-centric reconstruction and attention regularization are further proposed to increase the effectiveness of unlabeled data usage. Experiments are conducted on the various datasets in human and hand pose estimation tasks, especially using the cross-domain scenario. We demonstrated the effectiveness of ours by achieving the state-of-the-art accuracy on all datasets.
Authors: Georgii Gotin, Ekaterina Shumitskaya, Anastasia Antsiferova, Dmitriy Vatolin
Abstract: Recent studies have revealed that modern image and video quality assessment (IQA/VQA) metrics are vulnerable to adversarial attacks. An attacker can manipulate a video through preprocessing to artificially increase its quality score according to a certain metric, despite no actual improvement in visual quality. Most of the attacks studied in the literature are white-box attacks, while black-box attacks in the context of VQA have received less attention. Moreover, some research indicates a lack of transferability of adversarial examples generated for one model to another when applied to VQA. In this paper, we propose a cross-modal attack method, IC2VQA, aimed at exploring the vulnerabilities of modern VQA models. This approach is motivated by the observation that the low-level feature spaces of images and videos are similar. We investigate the transferability of adversarial perturbations across different modalities; specifically, we analyze how adversarial perturbations generated on a white-box IQA model with an additional CLIP module can effectively target a VQA model. The addition of the CLIP module serves as a valuable aid in increasing transferability, as the CLIP model is known for its effective capture of low-level semantics. Extensive experiments demonstrate that IC2VQA achieves a high success rate in attacking three black-box VQA models. We compare our method with existing black-box attack strategies, highlighting its superiority in terms of attack success within the same number of iterations and levels of attack strength. We believe that the proposed method will contribute to the deeper analysis of robust VQA metrics.
Authors: Sabri Mustafa Kahya, Boran Hamdi Sivrikaya, Muhammet Sami Yavuz, Eckehard Steinbach
Abstract: In this work, we propose a novel pipeline for face recognition and out-of-distribution (OOD) detection using short-range FMCW radar. The proposed system utilizes Range-Doppler and micro Range-Doppler Images. The architecture features a primary path (PP) responsible for the classification of in-distribution (ID) faces, complemented by intermediate paths (IPs) dedicated to OOD detection. The network is trained in two stages: first, the PP is trained using triplet loss to optimize ID face classification. In the second stage, the PP is frozen, and the IPs-comprising simple linear autoencoder networks-are trained specifically for OOD detection. Using our dataset generated with a 60 GHz FMCW radar, our method achieves an ID classification accuracy of 99.30% and an OOD detection AUROC of 96.91%.
Authors: Xu Li, Yi Zheng, Haotian Chen, Xiaolei Chen, Yuxuan Liang, Chenghang Lai
Abstract: Large Vision-Language Models (LVLMs) have achieved significant success in multimodal tasks by combining pre-trained vision encoders and large language models. However, current LVLMs mainly rely on features from the final layers of the vision encoder, neglecting complementary information in shallower layers. While recent methods have explored multi-layer features, they are often task-agnostic. We investigate the contributions of visual features from different encoder layers across 18 benchmarks and 6 task categories. Our results show that multi-layer features provide complementary strengths with varying task dependencies, and uniform fusion performs suboptimally. Based on these findings, we propose an instruction-guided vision aggregator that dynamically integrates multi-layer features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations show superior performance, and analysis reveals the dominance of mid-to-high-level features in semantic tasks and the critical role of low-level features in fine-grained perception. This work provides valuable insights into the adaptive use of hierarchical visual features in LVLMs, advancing more flexible multimodal systems.
Authors: Cesare Davide Pace, Alessandro Marco De Nunzio, Claudio De Stefano, Francesco Fontanella, Mario Molinara
Abstract: Human pose estimation, a vital task in computer vision, involves detecting and localising human joints in images and videos. While single-frame pose estimation has seen significant progress, it often fails to capture the temporal dynamics for understanding complex, continuous movements. We propose Poseidon, a novel multi-frame pose estimation architecture that extends the ViTPose model by integrating temporal information for enhanced accuracy and robustness to address these limitations. Poseidon introduces key innovations: (1) an Adaptive Frame Weighting (AFW) mechanism that dynamically prioritises frames based on their relevance, ensuring that the model focuses on the most informative data; (2) a Multi-Scale Feature Fusion (MSFF) module that aggregates features from different backbone layers to capture both fine-grained details and high-level semantics; and (3) a Cross-Attention module for effective information exchange between central and contextual frames, enhancing the model's temporal coherence. The proposed architecture improves performance in complex video scenarios and offers scalability and computational efficiency suitable for real-world applications. Our approach achieves state-of-the-art performance on the PoseTrack21 and PoseTrack18 datasets, achieving mAP scores of 88.3 and 87.8, respectively, outperforming existing methods.
Authors: Weichen Fan, Chenyang Si, Junhao Song, Zhenyu Yang, Yinan He, Long Zhuo, Ziqi Huang, Ziyue Dong, Jingwen He, Dongwei Pan, Yi Wang, Yuming Jiang, Yaohui Wang, Peng Gao, Xinyuan Chen, Hengjie Li, Dahua Lin, Yu Qiao, Ziwei Liu
Abstract: We present Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation. The overall Vchitect-2.0 system has several key designs. (1) By introducing a novel Multimodal Diffusion Block, our approach achieves consistent alignment between text descriptions and generated video frames, while maintaining temporal coherence across sequences. (2) To overcome memory and computational bottlenecks, we propose a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems. (3) Additionally, our enhanced data processing pipeline ensures the creation of Vchitect T2V DataVerse, a high-quality million-scale training dataset through rigorous annotation and aesthetic evaluation. Extensive benchmarking demonstrates that Vchitect-2.0 outperforms existing methods in video quality, training efficiency, and scalability, serving as a suitable base for high-fidelity video generation.
Authors: Mihai Masala, Marius Leordeanu
Abstract: In the current era of Machine Learning, Transformers have become the de facto approach across a variety of domains, such as computer vision and natural language processing. Transformer-based solutions are the backbone of current state-of-the-art methods for language generation, image and video classification, segmentation, action and object recognition, among many others. Interestingly enough, while these state-of-the-art methods produce impressive results in their respective domains, the problem of understanding the relationship between vision and language is still beyond our reach. In this work, we propose a common ground between vision and language based on events in space and time in an explainable and programmatic way, to connect learning-based vision and language state of the art models and provide a solution to the long standing problem of describing videos in natural language. We validate that our algorithmic approach is able to generate coherent, rich and relevant textual descriptions on videos collected from a variety of datasets, using both standard metrics (e.g. Bleu, ROUGE) and the modern LLM-as-a-Jury approach.
Authors: Ni Li, Ryan Jacobs, Matthew Lynch, Vidit Agrawal, Kevin Field, Dane Morgan
Abstract: Quantifying prediction uncertainty when applying object detection models to new, unlabeled datasets is critical in applied machine learning. This study introduces an approach to estimate the performance of deep learning-based object detection models for quantifying defects in transmission electron microscopy (TEM) images, focusing on detecting irradiation-induced cavities in TEM images of metal alloys. We developed a random forest regression model that predicts the object detection F1 score, a statistical metric used to evaluate the ability to accurately locate and classify objects of interest. The random forest model uses features extracted from the predictions of the object detection model whose uncertainty is being quantified, enabling fast prediction on new, unlabeled images. The mean absolute error (MAE) for predicting F1 of the trained model on test data is 0.09, and the $R^2$ score is 0.77, indicating there is a significant correlation between the random forest regression model predicted and true defect detection F1 scores. The approach is shown to be robust across three distinct TEM image datasets with varying imaging and material domains. Our approach enables users to estimate the reliability of a defect detection and segmentation model predictions and assess the applicability of the model to their specific datasets, providing valuable information about possible domain shifts and whether the model needs to be fine-tuned or trained on additional data to be maximally effective for the desired use case.
Authors: Zhengye Yang, Richard J. Radke
Abstract: We describe a method for modeling spatial context to enable video anomaly detection. The main idea is to discover regions that share similar object-level activities by clustering joint object attributes using Gaussian mixture models. We demonstrate that this straightforward approach, using orders of magnitude fewer parameters than competing models, achieves state-of-the-art performance in the challenging spatial-context-dependent Street Scene dataset. As a side benefit, the high-resolution discovered regions learned by the model also provide explainable normalcy maps for human operators without the need for any pre-trained segmentation model.
Authors: Md Meem Hossain, The Anh Han, Safina Showkat Ara, Zia Ush Shamszaman
Abstract: Human Activity Recognition (HAR) has gained significant importance with the growing use of sensor-equipped devices and large datasets. This paper evaluates the performance of three categories of models : classical machine learning, deep learning architectures, and Restricted Boltzmann Machines (RBMs) using five key benchmark datasets of HAR (UCI-HAR, OPPORTUNITY, PAMAP2, WISDM, and Berkeley MHAD). We assess various models, including Decision Trees, Random Forests, Convolutional Neural Networks (CNN), and Deep Belief Networks (DBNs), using metrics such as accuracy, precision, recall, and F1-score for a comprehensive comparison. The results show that CNN models offer superior performance across all datasets, especially on the Berkeley MHAD. Classical models like Random Forest do well on smaller datasets but face challenges with larger, more complex data. RBM-based models also show notable potential, particularly for feature learning. This paper offers a detailed comparison to help researchers choose the most suitable model for HAR tasks.
Authors: Isaac Corley, Simone Fobi Nsutezo, Anthony Ortiz, Caleb Robinson, Rahul Dodhia, Juan M. Lavista Ferres, Peyman Najafirad
Abstract: Remote sensing imagery is dense with objects and contextual visual information. There is a recent trend to combine paired satellite images and text captions for pretraining performant encoders for downstream tasks. However, while contrastive image-text methods like CLIP enable vision-language alignment and zero-shot classification ability, vision-only downstream performance tends to degrade compared to image-only pretraining, such as MAE. In this paper, we propose FLAVARS, a pretraining method that combines the best of both contrastive learning and masked modeling, along with geospatial alignment via contrastive location encoding. We find that FLAVARS significantly outperforms a baseline of SkyCLIP for vision-only tasks such as KNN classification and semantic segmentation, +6\% mIOU on SpaceNet1, while retaining the ability to perform zero-shot classification, unlike MAE pretrained methods.
Authors: Waqwoya Abebe, Sadegh Jafari, Sixing Yu, Akash Dutta, Jan Strube, Nathan R. Tallent, Luanzheng Guo, Pablo Munoz, Ali Jannesari
Abstract: Neural Architecture Search (NAS) is a powerful approach of automating the design of efficient neural architectures. In contrast to traditional NAS methods, recently proposed one-shot NAS methods prove to be more efficient in performing NAS. One-shot NAS works by generating a singular weight-sharing supernetwork that acts as a search space (container) of subnetworks. Despite its achievements, designing the one-shot search space remains a major challenge. In this work we propose a search space design strategy for Vision Transformer (ViT)-based architectures. In particular, we convert the Segment Anything Model (SAM) into a weight-sharing supernetwork called SuperSAM. Our approach involves automating the search space design via layer-wise structured pruning and parameter prioritization. While the structured pruning applies probabilistic removal of certain transformer layers, parameter prioritization performs weight reordering and slicing of MLP-blocks in the remaining layers. We train supernetworks on several datasets using the sandwich rule. For deployment, we enhance subnetwork discovery by utilizing a program autotuner to identify efficient subnetworks within the search space. The resulting subnetworks are 30-70% smaller in size compared to the original pre-trained SAM ViT-B, yet outperform the pretrained model. Our work introduces a new and effective method for ViT NAS search-space design.
Authors: Zhenyu Yu, Chee Seng Chan
Abstract: Generative AI presents transformative potential across various domains, from creative arts to scientific visualization. However, the utility of AI-generated imagery is often compromised by visual flaws, including anatomical inaccuracies, improper object placements, and misplaced textual elements. These imperfections pose significant challenges for practical applications. To overcome these limitations, we introduce \textit{Yuan}, a novel framework that autonomously corrects visual imperfections in text-to-image synthesis. \textit{Yuan} uniquely conditions on both the textual prompt and the segmented image, generating precise masks that identify areas in need of refinement without requiring manual intervention -- a common constraint in previous methodologies. Following the automated masking process, an advanced inpainting module seamlessly integrates contextually coherent content into the identified regions, preserving the integrity and fidelity of the original image and associated text prompts. Through extensive experimentation on publicly available datasets such as ImageNet100 and Stanford Dogs, along with a custom-generated dataset, \textit{Yuan} demonstrated superior performance in eliminating visual imperfections. Our approach consistently achieved higher scores in quantitative metrics, including NIQE, BRISQUE, and PI, alongside favorable qualitative evaluations. These results underscore \textit{Yuan}'s potential to significantly enhance the quality and applicability of AI-generated images across diverse fields.
Authors: Lizhi Chen, Zhong Qian, Peifeng Li, Qiaoming Zhu
Abstract: Multi-modal explanation involves the assessment of the veracity of a variety of different content, and relies on multiple information modalities to comprehensively consider the relevance and consistency between modalities. Most existing fake news video detection methods focus on improving accuracy while ignoring the importance of providing explanations. In this paper, we propose a novel problem - Fake News Video Explanation (FNVE) - Given a multimodal news containing both video and caption text, we aim to generate natural language explanations to reveal the truth of predictions. To this end, we develop FakeNVE, a new dataset of explanations for truthfully multimodal posts, where each explanation is a natural language (English) sentence describing the attribution of a news thread. We benchmark FakeNVE by using a multimodal transformer-based architecture. Subsequently, a BART-based autoregressive decoder is used as the generator. Empirical results show compelling results for various baselines (applicable to FNVE) across multiple evaluation metrics. We also perform human evaluation on explanation generation, achieving high scores for both adequacy and fluency.
Authors: Zelu Qi, Ping Shi, Shuqi Wang, Zhaoyang Zhang, Zefeng Ying, Da Pan
Abstract: Recent text-to-video (T2V) technology advancements, as demonstrated by models such as Gen3, Pika, and Sora, have significantly broadened its applicability and popularity. This progress has created a growing demand for accurate quality assessment metrics to evaluate the perceptual quality of text-generated videos and optimize video generation models. However, assessing the quality of text-generated videos remains challenging due to the presence of highly complex distortions, such as unnatural actions and phenomena that defy human cognition. To address these challenges, we constructed a large-scale benchmark dataset for \textbf{T}ext-generated \textbf{V}ideo \textbf{eval}uation, \textbf{T2VEval-Bench}, comprising 148 textual words and 1,783 videos generated by 12 models. During the subjective evaluation, we collected five key scores: overall impression, video quality, aesthetic quality, realness, and text-video consistency. For objective evaluation, we developed the \textbf{T2VEval} model, which assesses videos across three branches: quality, authenticity, and consistency. Using an attention-based fusion module, T2VEval effectively integrates features from each branch and predicts scores with the aid of a large oracle model. Additionally, we implemented a progressive training strategy, enabling each branch to learn targeted knowledge while maintaining synergy with the others. Experimental results demonstrate that T2VEval achieves state-of-the-art performance across multiple metrics. The dataset and code will be open-sourced upon completion of the follow-up work.
Authors: Sitong Gong, Yunzhi Zhuge, Lu Zhang, Zongxin Yang, Pingping Zhang, Huchuan Lu
Abstract: Existing methods for Video Reasoning Segmentation rely heavily on a single special token to represent the object in the keyframe or the entire video, inadequately capturing spatial complexity and inter-frame motion. To overcome these challenges, we propose VRS-HQ, an end-to-end video reasoning segmentation approach that leverages Multimodal Large Language Models (MLLMs) to inject rich spatiotemporal features into hierarchical tokens.Our key innovations include a Temporal Dynamic Aggregation (TDA) and a Token-driven Keyframe Selection (TKS). Specifically, we design frame-level
Authors: Runqi Wang, Sijie Xu, Tianyao He, Yang Chen, Wei Zhu, Dejia Song, Nemo Chen, Xu Tang, Yao Hu
Abstract: Face swapping transfers the identity of a source face to a target face while retaining the attributes like expression, pose, hair, and background of the target face. Advanced face swapping methods have achieved attractive results. However, these methods often inadvertently transfer identity information from the target face, compromising expression-related details and accurate identity. We propose a novel method DynamicFace that leverages the power of diffusion model and plug-and-play temporal layers for video face swapping. First, we introduce four fine-grained face conditions using 3D facial priors. All conditions are designed to be disentangled from each other for precise and unique control. Then, we adopt Face Former and ReferenceNet for high-level and detailed identity injection. Through experiments on the FF++ dataset, we demonstrate that our method achieves state-of-the-art results in face swapping, showcasing superior image quality, identity preservation, and expression accuracy. Besides, our method could be easily transferred to video domain with temporal attention layer. Our code and results will be available on the project page: https://dynamic-face.github.io/
Authors: Oscar Ramos-Soto, Jorge Ramos-Frutos, Ezequiel Perez-Zarate, Diego Oliva, Sandra E. Balderas-Mata
Abstract: Feature extraction techniques are crucial in medical image classification; however, classical feature extractors in addition to traditional machine learning classifiers often exhibit significant limitations in providing sufficient discriminative information for complex image sets. While Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) have shown promise in feature extraction, they are prone to overfitting due to the inherent characteristics of medical imaging data, including small sample sizes or high intra-class variance. In this work, the Medical Image Attention-based Feature Extractor (MIAFEx) is proposed, a novel method that employs a learnable refinement mechanism to enhance the classification token within the Transformer encoder architecture. This mechanism adjusts the token based on learned weights, improving the extraction of salient features and enhancing the model's adaptability to the challenges presented by medical imaging data. The MIAFEx output features quality is compared against classical feature extractors using traditional and hybrid classifiers. Also, the performance of these features is compared against modern CNN and ViT models in classification tasks, demonstrating its superiority in accuracy and robustness across multiple complex classification medical imaging datasets. This advantage is particularly pronounced in scenarios with limited training data, where traditional and modern models often struggle to generalize effectively. The source code of this proposal can be found at https://github.com/Oscar-RamosS/Medical-Image-Attention-based-Feature-Extractor-MIAFEx
URLs: https://github.com/Oscar-RamosS/Medical-Image-Attention-based-Feature-Extractor-MIAFEx
Authors: Leyuan Yang, Bailin Deng, Juyong Zhang
Abstract: Various SDF-based neural implicit surface reconstruction methods have been proposed recently, and have demonstrated remarkable modeling capabilities. However, due to the global nature and limited representation ability of a single network, existing methods still suffer from many drawbacks, such as limited accuracy and scale of the reconstruction. In this paper, we propose a versatile, scalable and high-quality neural implicit representation to address these issues. We integrate a divide-and-conquer approach into the neural SDF-based reconstruction. Specifically, we model the object or scene as a fusion of multiple independent local neural SDFs with overlapping regions. The construction of our representation involves three key steps: (1) constructing the distribution and overlap relationship of the local radiance fields based on object structure or data distribution, (2) relative pose registration for adjacent local SDFs, and (3) SDF blending. Thanks to the independent representation of each local region, our approach can not only achieve high-fidelity surface reconstruction, but also enable scalable scene reconstruction. Extensive experimental results demonstrate the effectiveness and practicality of our proposed method.
Authors: Jiaqi Huang, Zunnan Xu, Ting Liu, Yong Liu, Haonan Han, Kehong Yuan, Xiu Li
Abstract: In the domain of computer vision, Parameter-Efficient Tuning (PET) is increasingly replacing the traditional paradigm of pre-training followed by full fine-tuning. PET is particularly favored for its effectiveness in large foundation models, as it streamlines transfer learning costs and optimizes hardware utilization. However, the current PET methods are mainly designed for single-modal optimization. While some pioneering studies have undertaken preliminary explorations, they still remain at the level of aligned encoders (e.g., CLIP) and lack exploration of misaligned encoders. These methods show sub-optimal performance with misaligned encoders, as they fail to effectively align the multimodal features during fine-tuning. In this paper, we introduce DETRIS, a parameter-efficient tuning framework designed to enhance low-rank visual feature propagation by establishing dense interconnections between each layer and all preceding layers, which enables effective cross-modal feature interaction and adaptation to misaligned encoders. We also suggest using text adapters to improve textual features. Our simple yet efficient approach greatly surpasses state-of-the-art methods with 0.9% to 1.8% backbone parameter updates, evaluated on challenging benchmarks. Our project is available at \url{https://github.com/jiaqihuang01/DETRIS}.
Authors: Krishna Panthi
Abstract: This paper introduces a novel approach to enhance the performance of Gaussian Shading, a prevalent watermarking technique, by integrating the Exact Diffusion Inversion via Coupled Transformations (EDICT) framework. While Gaussian Shading traditionally embeds watermarks in a noise latent space, followed by iterative denoising for image generation and noise addition for watermark recovery, its inversion process is not exact, leading to potential watermark distortion. We propose to leverage EDICT's ability to derive exact inverse mappings to refine this process. Our method involves duplicating the watermark-infused noisy latent and employing a reciprocal, alternating denoising and noising scheme between the two latents, facilitated by EDICT. This allows for a more precise reconstruction of both the image and the embedded watermark. Empirical evaluation on standard datasets demonstrates that our integrated approach yields a slight, yet statistically significant improvement in watermark recovery fidelity. These results highlight the potential of EDICT to enhance existing diffusion-based watermarking techniques by providing a more accurate and robust inversion mechanism. To the best of our knowledge, this is the first work to explore the synergy between EDICT and Gaussian Shading for digital watermarking, opening new avenues for research in robust and high-fidelity watermark embedding and extraction.
Authors: Chenguang Liu, Yongchao Feng, Yanan Zhang, Qingjie Liu, Yunhong Wang
Abstract: In recent years, there has been significant advancement in object detection. However, applying off-the-shelf detectors to a new domain leads to significant performance drop, caused by the domain gap. These detectors exhibit higher-variance class-conditional distributions in the target domain than that in the source domain, along with mean shift. To address this problem, we propose the Prototype Augmented Compact Features (PACF) framework to regularize the distribution of intra-class features. Specifically, we provide an in-depth theoretical analysis on the lower bound of the target features-related likelihood and derive the prototype cross entropy loss to further calibrate the distribution of target RoI features. Furthermore, a mutual regularization strategy is designed to enable the linear and prototype-based classifiers to learn from each other, promoting feature compactness while enhancing discriminability. Thanks to this PACF framework, we have obtained a more compact cross-domain feature space, within which the variance of the target features' class-conditional distributions has significantly decreased, and the class-mean shift between the two domains has also been further reduced. The results on different adaptation settings are state-of-the-art, which demonstrate the board applicability and effectiveness of the proposed approach.
Authors: Kaleab A. Kinfu, Carolina Pacheco, Alice D. Sperry, Deana Crocetti, Bahar Tun\c{c}gen\c{c}, Stewart H. Mostofsky, Ren\'e Vidal
Abstract: Motor imitation impairments are commonly reported in individuals with autism spectrum conditions (ASCs), suggesting that motor imitation could be used as a phenotype for addressing autism heterogeneity. Traditional methods for assessing motor imitation are subjective, labor-intensive, and require extensive human training. Modern Computerized Assessment of Motor Imitation (CAMI) methods, such as CAMI-3D for motion capture data and CAMI-2D for video data, are less subjective. However, they rely on labor-intensive data normalization and cleaning techniques, and human annotations for algorithm training. To address these challenges, we propose CAMI-2DNet, a scalable and interpretable deep learning-based approach to motor imitation assessment in video data, which eliminates the need for data normalization, cleaning and annotation. CAMI-2DNet uses an encoder-decoder architecture to map a video to a motion encoding that is disentangled from nuisance factors such as body shape and camera views. To learn a disentangled representation, we employ synthetic data generated by motion retargeting of virtual characters through the reshuffling of motion, body shape, and camera views, as well as real participant data. To automatically assess how well an individual imitates an actor, we compute a similarity score between their motion encodings, and use it to discriminate individuals with ASCs from neurotypical (NT) individuals. Our comparative analysis demonstrates that CAMI-2DNet has a strong correlation with human scores while outperforming CAMI-2D in discriminating ASC vs NT children. Moreover, CAMI-2DNet performs comparably to CAMI-3D while offering greater practicality by operating directly on video data and without the need for ad-hoc data normalization and human annotations.
Authors: Giovanny Vazquez, Shengjie Zhai, Mei Yang
Abstract: Autonomous unmanned aerial vehicles (UAVs) integrated with edge computing capabilities empower real-time data processing directly on the device, dramatically reducing latency in critical scenarios such as wildfire detection. This study underscores Transfer Learning's (TL) significance in boosting the performance of object detectors for identifying wildfire smoke and flames, especially when trained on limited datasets, and investigates the impact TL has on edge computing metrics. With the latter focusing how TL-enhanced You Only Look Once (YOLO) models perform in terms of inference time, power usage, and energy consumption when using edge computing devices. This study utilizes the Aerial Fire and Smoke Essential (AFSE) dataset as the target, with the Flame and Smoke Detection Dataset (FASDD) and the Microsoft Common Objects in Context (COCO) dataset serving as source datasets. We explore a two-stage cascaded TL method, utilizing D-Fire or FASDD as initial stage target datasets and AFSE as the subsequent stage. Through fine-tuning, TL significantly enhances detection precision, achieving up to 79.2% mean Average Precision (mAP@0.5), reduces training time, and increases model generalizability across the AFSE dataset. However, cascaded TL yielded no notable improvements and TL alone did not benefit the edge computing metrics evaluated. Lastly, this work found that YOLOv5n remains a powerful model when lacking hardware acceleration, finding that YOLOv5n can process images nearly twice as fast as its newer counterpart, YOLO11n. Overall, the results affirm TL's role in augmenting the accuracy of object detectors while also illustrating that additional enhancements are needed to improve edge computing performance.
Authors: Junda Cheng, Longliang Liu, Gangwei Xu, Xianqi Wang, Zhaoxing Zhang, Yong Deng, Jinliang Zang, Yurui Chen, Zhipeng Cai, Xin Yang
Abstract: Stereo matching recovers depth from image correspondences. Existing methods struggle to handle ill-posed regions with limited matching cues, such as occlusions and textureless areas. To address this, we propose MonSter, a novel method that leverages the complementary strengths of monocular depth estimation and stereo matching. MonSter integrates monocular depth and stereo matching into a dual-branch architecture to iteratively improve each other. Confidence-based guidance adaptively selects reliable stereo cues for monodepth scale-shift recovery. The refined monodepth is in turn guides stereo effectively at ill-posed regions. Such iterative mutual enhancement enables MonSter to evolve monodepth priors from coarse object-level structures to pixel-level geometry, fully unlocking the potential of stereo matching. As shown in Fig.1, MonSter ranks 1st across five most commonly used leaderboards -- SceneFlow, KITTI 2012, KITTI 2015, Middlebury, and ETH3D. Achieving up to 49.5% improvements (Bad 1.0 on ETH3D) over the previous best method. Comprehensive analysis verifies the effectiveness of MonSter in ill-posed regions. In terms of zero-shot generalization, MonSter significantly and consistently outperforms state-of-the-art across the board. The code is publicly available at: https://github.com/Junda24/MonSter.
Authors: Xinya Ji, Gaspard Zoss, Prashanth Chandran, Lingchen Yang, Xun Cao, Barbara Solenthaler, Derek Bradley
Abstract: 2D portrait animation has experienced significant advancements in recent years. Much research has utilized the prior knowledge embedded in large generative diffusion models to enhance high-quality image manipulation. However, most methods only focus on generating RGB images as output, and the co-generation of consistent visual plus 3D output remains largely under-explored. In our work, we propose to jointly learn the visual appearance and depth simultaneously in a diffusion-based portrait image generator. Our method embraces the end-to-end diffusion paradigm and introduces a new architecture suitable for learning this conditional joint distribution, consisting of a reference network and a channel-expanded diffusion backbone. Once trained, our framework can be efficiently adapted to various downstream applications, such as facial depth-to-image and image-to-depth generation, portrait relighting, and audio-driven talking head animation with consistent 3D output.
Authors: Xianqi Wang, Hao Yang, Gangwei Xu, Junda Cheng, Min Lin, Yong Deng, Jinliang Zang, Yurui Chen, Xin Yang
Abstract: State-of-the-art supervised stereo matching methods have achieved amazing results on various benchmarks. However, these data-driven methods suffer from generalization to real-world scenarios due to the lack of real-world annotated data. In this paper, we propose StereoGen, a novel pipeline for high-quality stereo image generation. This pipeline utilizes arbitrary single images as left images and pseudo disparities generated by a monocular depth estimation model to synthesize high-quality corresponding right images. Unlike previous methods that fill the occluded area in warped right images using random backgrounds or using convolutions to take nearby pixels selectively, we fine-tune a diffusion inpainting model to recover the background. Images generated by our model possess better details and undamaged semantic structures. Besides, we propose Training-free Confidence Generation and Adaptive Disparity Selection. The former suppresses the negative effect of harmful pseudo ground truth during stereo training, while the latter helps generate a wider disparity distribution and better synthetic images. Experiments show that models trained under our pipeline achieve state-of-the-art zero-shot generalization results among all published methods. The code will be available upon publication of the paper.
Authors: Dongzhihan Wang, Yang Yang, Liang Xu
Abstract: Visual odometry (VO) plays a crucial role in autonomous driving, robotic navigation, and other related tasks by estimating the position and orientation of a camera based on visual input. Significant progress has been made in data-driven VO methods, particularly those leveraging deep learning techniques to extract image features and estimate camera poses. However, these methods often struggle in low-light conditions because of the reduced visibility of features and the increased difficulty of matching keypoints. To address this limitation, we introduce BrightVO, a novel VO model based on Transformer architecture, which not only performs front-end visual feature extraction, but also incorporates a multi-modality refinement module in the back-end that integrates Inertial Measurement Unit (IMU) data. Using pose graph optimization, this module iteratively refines pose estimates to reduce errors and improve both accuracy and robustness. Furthermore, we create a synthetic low-light dataset, KiC4R, which includes a variety of lighting conditions to facilitate the training and evaluation of VO frameworks in challenging environments. Experimental results demonstrate that BrightVO achieves state-of-the-art performance on both the KiC4R dataset and the KITTI benchmarks. Specifically, it provides an average improvement of 20% in pose estimation accuracy in normal outdoor environments and 259% in low-light conditions, outperforming existing methods. For widespread use and further development, the research work is fully open-source at https://github.com/Anastasiawd/BrightVO.
Authors: Chen Chen, Mengyuan Sun, Xueluan Gong, Yanjiao Chen, Qian Wang
Abstract: Facial recognition models are increasingly employed by commercial enterprises, government agencies, and cloud service providers for identity verification, consumer services, and surveillance. These models are often trained using vast amounts of facial data processed and stored in cloud-based platforms, raising significant privacy concerns. Users' facial images may be exploited without their consent, leading to potential data breaches and misuse. This survey presents a comprehensive review of current methods aimed at preserving facial image privacy in cloud-based services. We categorize these methods into two primary approaches: image obfuscation-based protection and adversarial perturbation-based protection. We provide an in-depth analysis of both categories, offering qualitative and quantitative comparisons of their effectiveness. Additionally, we highlight unresolved challenges and propose future research directions to improve privacy preservation in cloud computing environments.
Authors: Anant Khandelwal
Abstract: Animating clipart images with seamless motion while maintaining visual fidelity and temporal coherence presents significant challenges. Existing methods, such as AniClipart, effectively model spatial deformations but often fail to ensure smooth temporal transitions, resulting in artifacts like abrupt motions and geometric distortions. Similarly, text-to-video (T2V) and image-to-video (I2V) models struggle to handle clipart due to the mismatch in statistical properties between natural video and clipart styles. This paper introduces FlexiClip, a novel approach designed to overcome these limitations by addressing the intertwined challenges of temporal consistency and geometric integrity. FlexiClip extends traditional B\'ezier curve-based trajectory modeling with key innovations: temporal Jacobians to correct motion dynamics incrementally, continuous-time modeling via probability flow ODEs (pfODEs) to mitigate temporal noise, and a flow matching loss inspired by GFlowNet principles to optimize smooth motion transitions. These enhancements ensure coherent animations across complex scenarios involving rapid movements and non-rigid deformations. Extensive experiments validate the effectiveness of FlexiClip in generating animations that are not only smooth and natural but also structurally consistent across diverse clipart types, including humans and animals. By integrating spatial and temporal modeling with pre-trained video diffusion models, FlexiClip sets a new standard for high-quality clipart animation, offering robust performance across a wide range of visual content. Project Page: https://creative-gen.github.io/flexiclip.github.io/
Authors: Siqi Li, Zhengkai Jiang, Jiawei Zhou, Zhihong Liu, Xiaowei Chi, Haoqian Wang
Abstract: Virtual try-on has emerged as a pivotal task at the intersection of computer vision and fashion, aimed at digitally simulating how clothing items fit on the human body. Despite notable progress in single-image virtual try-on (VTO), current methodologies often struggle to preserve a consistent and authentic appearance of clothing across extended video sequences. This challenge arises from the complexities of capturing dynamic human pose and maintaining target clothing characteristics. We leverage pre-existing video foundation models to introduce RealVVT, a photoRealistic Video Virtual Try-on framework tailored to bolster stability and realism within dynamic video contexts. Our methodology encompasses a Clothing & Temporal Consistency strategy, an Agnostic-guided Attention Focus Loss mechanism to ensure spatial consistency, and a Pose-guided Long Video VTO technique adept at handling extended video sequences.Extensive experiments across various datasets confirms that our approach outperforms existing state-of-the-art models in both single-image and video VTO tasks, offering a viable solution for practical applications within the realms of fashion e-commerce and virtual fitting environments.
Authors: Jaemyung Yu, Jaehyun Choi, Dong-Jae Lee, HyeongGwon Hong, Junmo Kim
Abstract: Unsupervised representation learning has significantly advanced various machine learning tasks. In the computer vision domain, state-of-the-art approaches utilize transformations like random crop and color jitter to achieve invariant representations, embedding semantically the same inputs despite transformations. However, this can degrade performance in tasks requiring precise features, such as localization or flower classification. To address this, recent research incorporates equivariant representation learning, which captures transformation-sensitive information. However, current methods depend on transformation labels and thus struggle with interdependency and complex transformations. We propose Self-supervised Transformation Learning (STL), replacing transformation labels with transformation representations derived from image pairs. The proposed method ensures transformation representation is image-invariant and learns corresponding equivariant transformations, enhancing performance without increased batch complexity. We demonstrate the approach's effectiveness across diverse classification and detection tasks, outperforming existing methods in 7 out of 11 benchmarks and excelling in detection. By integrating complex transformations like AugMix, unusable by prior equivariant methods, this approach enhances performance across tasks, underscoring its adaptability and resilience. Additionally, its compatibility with various base models highlights its flexibility and broad applicability. The code is available at https://github.com/jaemyung-u/stl.
Authors: Shiyu Wu, Jing Liu, Jing Li, Yequan Wang
Abstract: Current fake image detectors trained on large synthetic image datasets perform satisfactorily on limited studied generative models. However, they suffer a notable performance decline over unseen models. Besides, collecting adequate training data from online generative models is often expensive or infeasible. To overcome these issues, we propose Few-Shot Detector (FSD), a novel AI-generated image detector which learns a specialized metric space to effectively distinguish unseen fake images by utilizing very few samples. Experiments show FSD achieves state-of-the-art performance by $+7.4\%$ average ACC on GenImage dataset. More importantly, our method is better capable of capturing the intra-category common features in unseen images without further training.
Authors: Haopeng Li, Tom Drummond, Mingming Gong, Mohammed Bennamoun, Qiuhong Ke
Abstract: Significant progress has been made in the field of video question answering (VideoQA) thanks to deep learning and large-scale pretraining. Despite the presence of sophisticated model structures and powerful video-text foundation models, most existing methods focus solely on maximizing the correlation between answers and video-question pairs during training. We argue that these models often establish shortcuts, resulting in spurious correlations between questions and answers, especially when the alignment between video and text data is suboptimal. To address these spurious correlations, we propose a novel training framework in which the model is compelled to acknowledge its ignorance when presented with an intervened question, rather than making guesses solely based on superficial question-answer correlations. We introduce methodologies for intervening in questions, utilizing techniques such as displacement and perturbation, and design frameworks for the model to admit its lack of knowledge in both multi-choice VideoQA and open-ended settings. In practice, we integrate a state-of-the-art model into our framework to validate its effectiveness. The results clearly demonstrate that our framework can significantly enhance the performance of VideoQA models with minimal structural modifications.
Authors: Alain Komaty, Hatef Otroshi Shahreza, Anjith George, Sebastien Marcel
Abstract: This study highlights the potential of ChatGPT (specifically GPT-4o) as a competitive alternative for Face Presentation Attack Detection (PAD), outperforming several PAD models, including commercial solutions, in specific scenarios. Our results show that GPT-4o demonstrates high consistency, particularly in few-shot in-context learning, where its performance improves as more examples are provided (reference data). We also observe that detailed prompts enable the model to provide scores reliably, a behavior not observed with concise prompts. Additionally, explanation-seeking prompts slightly enhance the model's performance by improving its interpretability. Remarkably, the model exhibits emergent reasoning capabilities, correctly predicting the attack type (print or replay) with high accuracy in few-shot scenarios, despite not being explicitly instructed to classify attack types. Despite these strengths, GPT-4o faces challenges in zero-shot tasks, where its performance is limited compared to specialized PAD systems. Experiments were conducted on a subset of the SOTERIA dataset, ensuring compliance with data privacy regulations by using only data from consenting individuals. These findings underscore GPT-4o's promise in PAD applications, laying the groundwork for future research to address broader data privacy concerns and improve cross-dataset generalization. Code available here: https://gitlab.idiap.ch/bob/bob.paper.wacv2025_chatgpt_face_pad
URLs: https://gitlab.idiap.ch/bob/bob.paper.wacv2025_chatgpt_face_pad
Authors: Isaac Osei Agyemanga, Liaoyuan Zeng, Jianwen Chena, Isaac Adjei-Mensah, Daniel Acheampong
Abstract: Accurate detection and resilience of object detectors in structural damage detection are important in ensuring the continuous use of civil infrastructure. However, achieving robustness in object detectors remains a persistent challenge, impacting their ability to generalize effectively. This study proposes DetectorX, a robust framework for structural damage detection coupled with a micro drone. DetectorX addresses the challenges of object detector robustness by incorporating two innovative modules: a stem block and a spiral pooling technique. The stem block introduces a dynamic visual modality by leveraging the outputs of two Deep Convolutional Neural Network (DCNN) models. The framework employs the proposed event-based reward reinforcement learning to constrain the actions of a parent and child DCNN model leading to a reward. This results in the induction of two dynamic visual modalities alongside the Red, Green, and Blue (RGB) data. This enhancement significantly augments DetectorX's perception and adaptability in diverse environmental situations. Further, a spiral pooling technique, an online image augmentation method, strengthens the framework by increasing feature representations by concatenating spiraled and average/max pooled features. In three extensive experiments: (1) comparative and (2) robustness, which use the Pacific Earthquake Engineering Research Hub ImageNet dataset, and (3) field-experiment, DetectorX performed satisfactorily across varying metrics, including precision (0.88), recall (0.84), average precision (0.91), mean average precision (0.76), and mean average recall (0.73), compared to the competing detectors including You Only Look Once X-medium (YOLOX-m) and others. The study's findings indicate that DetectorX can provide satisfactory results and demonstrate resilience in challenging environments.
Authors: Matej Suchanek, Miroslav Purkrabek, Jiri Matas
Abstract: UV map estimation is used in computer vision for detailed analysis of human posture or activity. Previous methods assign pixels to body model vertices by comparing pixel descriptors independently, without enforcing global coherence or plausibility in the UV map. We propose Pose-Constrained Continuous Surface Embeddings (PC-CSE), which integrates estimated 2D human pose into the pixel-to-vertex assignment process. The pose provides global anatomical constraints, ensuring that UV maps remain coherent while preserving local precision. Evaluation on DensePose COCO demonstrates consistent improvement, regardless of the chosen 2D human pose model. Whole-body poses offer better constraints by incorporating additional details about the hands and feet. Conditioning UV maps with human pose reduces invalid mappings and enhances anatomical plausibility. In addition, we highlight inconsistencies in the ground-truth annotations.
Authors: Zhipeng Ye, Feng Jiang, Qiufeng Wang, Kaizhu Huang, Jiaqi Huang
Abstract: CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g. zero- or few-shot classification) is a hot topic in multimodal learning. However, current studies primarily focus on either prompt learning for text or adapter tuning for vision, without fully exploiting the complementary information and correlations among image-text pairs. In this paper, we propose an Image Description Enhanced CLIP-Adapter (IDEA) method to adapt CLIP to few-shot image classification tasks. This method captures fine-grained features by leveraging both visual features and textual descriptions of images. IDEA is a training-free method for CLIP, and it can be comparable to or even exceeds state-of-the-art models on multiple tasks. Furthermore, we introduce Trainable-IDEA (T-IDEA), which extends IDEA by adding two lightweight learnable components (i.e., a projector and a learnable latent space), further enhancing the model's performance and achieving SOTA results on 11 datasets. As one important contribution, we employ the Llama model and design a comprehensive pipeline to generate textual descriptions for images of 11 datasets, resulting in a total of 1,637,795 image-text pairs, named "IMD-11". Our code and data are released at https://github.com/FourierAI/IDEA.
Authors: Olga Zatsarynna, Emad Bahrami, Yazan Abu Farha, Gianpiero Francesca, Juergen Gall
Abstract: Our work addresses the problem of stochastic long-term dense anticipation. The goal of this task is to predict actions and their durations several minutes into the future based on provided video observations. Anticipation over extended horizons introduces high uncertainty, as a single observation can lead to multiple plausible future outcomes. To address this uncertainty, stochastic models are designed to predict several potential future action sequences. Recent work has further proposed to incorporate uncertainty modelling for observed frames by simultaneously predicting per-frame past and future actions in a unified manner. While such joint modelling of actions is beneficial, it requires long-range temporal capabilities to connect events across distant past and future time points. However, the previous work struggles to achieve such a long-range understanding due to its limited and/or sparse receptive field. To alleviate this issue, we propose a novel MANTA (MAmba for ANTicipation) network. Our model enables effective long-term temporal modelling even for very long sequences while maintaining linear complexity in sequence length. We demonstrate that our approach achieves state-of-the-art results on three datasets - Breakfast, 50Salads, and Assembly101 - while also significantly improving computational and memory efficiency.
Authors: Tengpeng Li, Hanli Wang, Xianfei Li, Wenlong Liao, Tao He, Pai Peng
Abstract: Autonomous driving is a challenging task that requires perceiving and understanding the surrounding environment for safe trajectory planning. While existing vision-based end-to-end models have achieved promising results, these methods are still facing the challenges of vision understanding, decision reasoning and scene generalization. To solve these issues, a generative planning with 3D-vision language pre-training model named GPVL is proposed for end-to-end autonomous driving. The proposed paradigm has two significant aspects. On one hand, a 3D-vision language pre-training module is designed to bridge the gap between visual perception and linguistic understanding in the bird's eye view. On the other hand, a cross-modal language model is introduced to generate holistic driving decisions and fine-grained trajectories with perception and navigation information in an auto-regressive manner. Experiments on the challenging nuScenes dataset demonstrate that the proposed scheme achieves excellent performances compared with state-of-the-art methods. Besides, the proposed GPVL presents strong generalization ability and real-time potential when handling high-level commands in various scenarios. It is believed that the effective, robust and efficient performance of GPVL is crucial for the practical application of future autonomous driving systems. Code is available at https://github.com/ltp1995/GPVL
Authors: Jhe-Hao Lin, Yi Yao, Chan-Feng Hsu, Hongxia Xie, Hong-Han Shuai, Wen-Huang Cheng
Abstract: Knowledge distillation (KD) involves transferring knowledge from a pre-trained heavy teacher model to a lighter student model, thereby reducing the inference cost while maintaining comparable effectiveness. Prior KD techniques typically assume homogeneity between the teacher and student models. However, as technology advances, a wide variety of architectures have emerged, ranging from initial Convolutional Neural Networks (CNNs) to Vision Transformers (ViTs), and Multi-Level Perceptrons (MLPs). Consequently, developing a universal KD framework compatible with any architecture has become an important research topic. In this paper, we introduce a feature-based one-for-all (FOFA) KD framework to enable feature distillation across diverse architecture. Our framework comprises two key components. First, we design prompt tuning blocks that incorporate student feedback, allowing teacher features to adapt to the student model's learning process. Second, we propose region-aware attention to mitigate the view mismatch problem between heterogeneous architecture. By leveraging these two modules, effective distillation of intermediate features can be achieved across heterogeneous architectures. Extensive experiments on CIFAR, ImageNet, and COCO demonstrate the superiority of the proposed method.
Authors: Hao Tang, Ling Shao, Nicu Sebe, Luc Van Gool
Abstract: In this paper, we propose a novel cross-attention-based generative adversarial network (GAN) for the challenging person image generation task. Cross-attention is a novel and intuitive multi-modal fusion method in which an attention/correlation matrix is calculated between two feature maps of different modalities. Specifically, we propose the novel XingGAN (or CrossingGAN), which consists of two generation branches that capture the person's appearance and shape, respectively. Moreover, we propose two novel cross-attention blocks to effectively transfer and update the person's shape and appearance embeddings for mutual improvement. This has not been considered by any other existing GAN-based image generation work. To further learn the long-range correlations between different person poses at different scales and sub-regions, we propose two novel multi-scale cross-attention blocks. To tackle the issue of independent correlation computations within the cross-attention mechanism leading to noisy and ambiguous attention weights, which hinder performance improvements, we propose a module called enhanced attention (EA). Lastly, we introduce a novel densely connected co-attention module to fuse appearance and shape features at different stages effectively. Extensive experiments on two public datasets demonstrate that the proposed method outperforms current GAN-based methods and performs on par with diffusion-based methods. However, our method is significantly faster than diffusion-based methods in both training and inference.
Authors: Gabriella Pangelinan, Grace Bezold, Haiyu Wu, Michael C. King, Kevin W. Bowyer
Abstract: Facial brightness is a key image quality factor impacting face recognition accuracy differentials across demographic groups. In this work, we aim to decrease the accuracy gap between the similarity score distributions for Caucasian and African American female mated image pairs, as measured by d' between distributions. To balance brightness across demographic groups, we conduct three experiments, interpreting brightness in the face skin region either as median pixel value or as the distribution of pixel values. Balancing based on median brightness alone yields up to a 46.8% decrease in d', while balancing based on brightness distribution yields up to a 57.6% decrease. In all three cases, the similarity scores of the individual distributions improve, with mean scores maximally improving 5.9% for Caucasian females and 3.7% for African American females.
Authors: Sadia Afrin Rimi, Md. Jalal Uddin Chowdhury, Rifat Abdullah, Iftekhar Ahmed, Mahrima Akter Mim, Mohammad Shoaib Rahman
Abstract: The number of people living in this agricultural nation of ours, which is surrounded by lush greenery, is growing on a daily basis. As a result of this, the level of arable land is decreasing, as well as residential houses and industrial factories. The food crisis is becoming the main threat for us in the upcoming days. Because on the one hand, the population is increasing, and on the other hand, the amount of food crop production is decreasing due to the attack of diseases. Rice is one of the most significant cultivated crops since it provides food for more than half of the world's population. Bangladesh is dependent on rice (Oryza sativa) as a vital crop for its agriculture, but it faces a significant problem as a result of the ongoing decline in rice yield brought on by common diseases. Early disease detection is the main difficulty in rice crop cultivation. In this paper, we proposed our own dataset, which was collected from the Bangladesh field, and also applied deep learning and transfer learning models for the evaluation of the datasets. We elaborately explain our dataset and also give direction for further research work to serve society using this dataset. We applied a light CNN model and pre-trained InceptionNet-V2, EfficientNet-V2, and MobileNet-V2 models, which achieved 91.5% performance for the EfficientNet-V2 model of this work. The results obtained assaulted other models and even exceeded approaches that are considered to be part of the state of the art. It has been demonstrated by this study that it is possible to precisely and effectively identify diseases that affect rice leaves using this unbiased datasets. After analysis of the performance of different models, the proposed datasets are significant for the society for research work to provide solutions for decreasing rice leaf disease.
Authors: Benoit Brummer, Christophe De Vleeschouwer
Abstract: This paper introduces the Raw Natural Image Noise Dataset (RawNIND), a diverse collection of paired raw images designed to support the development of denoising models that generalize across sensors, image development workflows, and styles. Two denoising methods are proposed: one operates directly on raw Bayer data, leveraging computational efficiency, while the other processes linear RGB images for improved generalization to different sensors, with both preserving flexibility for subsequent development. Both methods outperform traditional approaches which rely on developed images. Additionally, the integration of denoising and compression at the raw data level significantly enhances rate-distortion performance and computational efficiency. These findings suggest a paradigm shift toward raw data workflows for efficient and flexible image processing.
Authors: Javier Rodriguez-Juan, David Ortiz-Perez, Manuel Benavent-Lledo, David Mulero-P\'erez, Pablo Ruiz-Ponce, Adrian Orihuela-Torres, Jose Garcia-Rodriguez, Esther Sebasti\'an-Gonz\'alez
Abstract: The current biodiversity loss crisis makes animal monitoring a relevant field of study. In light of this, data collected through monitoring can provide essential insights, and information for decision-making aimed at preserving global biodiversity. Despite the importance of such data, there is a notable scarcity of datasets featuring videos of birds, and none of the existing datasets offer detailed annotations of bird behaviors in video format. In response to this gap, our study introduces the first fine-grained video dataset specifically designed for bird behavior detection and species classification. This dataset addresses the need for comprehensive bird video datasets and provides detailed data on bird actions, facilitating the development of deep learning models to recognize these, similar to the advancements made in human action recognition. The proposed dataset comprises 178 videos recorded in Spanish wetlands, capturing 13 different bird species performing 7 distinct behavior classes. In addition, we also present baseline results using state of the art models on two tasks: bird behavior recognition and species classification.
Authors: Francisco Mauro, Emanoel Thyago, Othon Vinicius, Rodrigo Abreu, Kelvin Cunha, Jos\'e Gabriel, Rafael Barros, Thales Bezerra, Manoel Henriques, Natalia Lopes, \'Erico Moutinho, J\'essica Guido, Tsang Ing Ren, Paulo Borba
Abstract: AI algorithms have become valuable in aiding professionals in healthcare. The increasing confidence obtained by these models is helpful in critical decision demands. In clinical dermatology, classification models can detect malignant lesions on patients' skin using only RGB images as input. However, most learning-based methods employ data acquired from dermoscopic datasets on training, which are large and validated by a gold standard. Clinical models aim to deal with classification on users' smartphone cameras that do not contain the corresponding resolution provided by dermoscopy. Also, clinical applications bring new challenges. It can contain captures from uncontrolled environments, skin tone variations, viewpoint changes, noises in data and labels, and unbalanced classes. A possible alternative would be to use transfer learning to deal with the clinical images. However, as the number of samples is low, it can cause degradations on the model's performance; the source distribution used in training differs from the test set. This work aims to evaluate the gap between dermoscopic and clinical samples and understand how the dataset variations impact training. It assesses the main differences between distributions that disturb the model's prediction. Finally, from experiments on different architectures, we argue how to combine the data from divergent distributions, decreasing the impact on the model's final accuracy.
Authors: Qi Ma, Runyi Yang, Bin Ren, Ender Konukoglu, Luc Van Gool, Danda Pani Paudel
Abstract: Localizing text descriptions in large-scale 3D scenes is inherently an ambiguous task. This nonetheless arises while describing general concepts, e.g. all traffic lights in a city. To facilitate reasoning based on such concepts, text localization in the form of distribution is required. In this paper, we generate the distribution of the camera poses conditioned upon the textual description. To facilitate such generation, we propose a diffusion-based architecture that conditionally diffuses the noisy 6DoF camera poses to their plausible locations. The conditional signals are derived from the text descriptions, using the pre-trained text encoders. The connection between text descriptions and pose distribution is established through pretrained Vision-Language-Model, i.e. CLIP. Furthermore, we demonstrate that the candidate poses for the distribution can be further refined by rendering potential poses using 3D Gaussian splatting, guiding incorrectly posed samples towards locations that better align with the textual description, through visual reasoning. We demonstrate the effectiveness of our method by comparing it with both standard retrieval methods and learning-based approaches. Our proposed method consistently outperforms these baselines across all five large-scale datasets. Our source code and dataset will be made publicly available.
Authors: Haozhe Xie, Zhaoxi Chen, Fangzhou Hong, Ziwei Liu
Abstract: 3D scene generation has garnered growing attention in recent years and has made significant progress. Generating 4D cities is more challenging than 3D scenes due to the presence of structurally complex, visually diverse objects like buildings and vehicles, and heightened human sensitivity to distortions in urban environments. To tackle these issues, we propose CityDreamer4D, a compositional generative model specifically tailored for generating unbounded 4D cities. Our main insights are 1) 4D city generation should separate dynamic objects (e.g., vehicles) from static scenes (e.g., buildings and roads), and 2) all objects in the 4D scene should be composed of different types of neural fields for buildings, vehicles, and background stuff. Specifically, we propose Traffic Scenario Generator and Unbounded Layout Generator to produce dynamic traffic scenarios and static city layouts using a highly compact BEV representation. Objects in 4D cities are generated by combining stuff-oriented and instance-oriented neural fields for background stuff, buildings, and vehicles. To suit the distinct characteristics of background stuff and instances, the neural fields employ customized generative hash grids and periodic positional embeddings as scene parameterizations. Furthermore, we offer a comprehensive suite of datasets for city generation, including OSM, GoogleEarth, and CityTopia. The OSM dataset provides a variety of real-world city layouts, while the Google Earth and CityTopia datasets deliver large-scale, high-quality city imagery complete with 3D instance annotations. Leveraging its compositional design, CityDreamer4D supports a range of downstream applications, such as instance editing, city stylization, and urban simulation, while delivering state-of-the-art performance in generating realistic 4D cities.
Authors: Chenyang Si, Weichen Fan, Zhengyao Lv, Ziqi Huang, Yu Qiao, Ziwei Liu
Abstract: Video generation has achieved remarkable progress with the introduction of diffusion models, which have significantly improved the quality of generated videos. However, recent research has primarily focused on scaling up model training, while offering limited insights into the direct impact of representations on the video generation process. In this paper, we initially investigate the characteristics of features in intermediate layers, finding substantial variations in attention maps across different layers. These variations lead to unstable semantic representations and contribute to cumulative differences between features, which ultimately reduce the similarity between adjacent frames and negatively affect temporal coherence. To address this, we propose RepVideo, an enhanced representation framework for text-to-video diffusion models. By accumulating features from neighboring layers to form enriched representations, this approach captures more stable semantic information. These enhanced representations are then used as inputs to the attention mechanism, thereby improving semantic expressiveness while ensuring feature consistency across adjacent frames. Extensive experiments demonstrate that our RepVideo not only significantly enhances the ability to generate accurate spatial appearances, such as capturing complex spatial relationships between multiple objects, but also improves temporal consistency in video generation.
Authors: Aditya Bhat, Rupak Bose, Chinedu Innocent Nwoye, Nicolas Padoy
Abstract: Acquiring and annotating surgical data is often resource-intensive, ethical constraining, and requiring significant expert involvement. While generative AI models like text-to-image can alleviate data scarcity, incorporating spatial annotations, such as segmentation masks, is crucial for precision-driven surgical applications, simulation, and education. This study introduces both a novel task and method, SimGen, for Simultaneous Image and Mask Generation. SimGen is a diffusion model based on the DDPM framework and Residual U-Net, designed to jointly generate high-fidelity surgical images and their corresponding segmentation masks. The model leverages cross-correlation priors to capture dependencies between continuous image and discrete mask distributions. Additionally, a Canonical Fibonacci Lattice (CFL) is employed to enhance class separability and uniformity in the RGB space of the masks. SimGen delivers high-fidelity images and accurate segmentation masks, outperforming baselines across six public datasets assessed on image and semantic inception distance metrics. Ablation study shows that the CFL improves mask quality and spatial separation. Downstream experiments suggest generated image-mask pairs are usable if regulations limit human data release for research. This work offers a cost-effective solution for generating paired surgical images and complex labels, advancing surgical AI development by reducing the need for expensive manual annotations.
Authors: Ruixiang Jiang, Changwen Chen
Abstract: We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
Authors: Jingyuan Chen, Fuchen Long, Jie An, Zhaofan Qiu, Ting Yao, Jiebo Luo, Tao Mei
Abstract: The first-in-first-out (FIFO) video diffusion, built on a pre-trained text-to-video model, has recently emerged as an effective approach for tuning-free long video generation. This technique maintains a queue of video frames with progressively increasing noise, continuously producing clean frames at the queue's head while Gaussian noise is enqueued at the tail. However, FIFO-Diffusion often struggles to keep long-range temporal consistency in the generated videos due to the lack of correspondence modeling across frames. In this paper, we propose Ouroboros-Diffusion, a novel video denoising framework designed to enhance structural and content (subject) consistency, enabling the generation of consistent videos of arbitrary length. Specifically, we introduce a new latent sampling technique at the queue tail to improve structural consistency, ensuring perceptually smooth transitions among frames. To enhance subject consistency, we devise a Subject-Aware Cross-Frame Attention (SACFA) mechanism, which aligns subjects across frames within short segments to achieve better visual coherence. Furthermore, we introduce self-recurrent guidance. This technique leverages information from all previous cleaner frames at the front of the queue to guide the denoising of noisier frames at the end, fostering rich and contextual global information interaction. Extensive experiments of long video generation on the VBench benchmark demonstrate the superiority of our Ouroboros-Diffusion, particularly in terms of subject consistency, motion smoothness, and temporal consistency.
Authors: Kuanren Qian, Genesis Omana Suarez, Toshihiko Nambara, Takahisa Kanekiyo, Yongjie Jessica Zhang
Abstract: Neurodevelopmental disorders (NDDs) cover a variety of conditions, including autism spectrum disorder, attention-deficit/hyperactivity disorder, and epilepsy, which impair the central and peripheral nervous systems. Their high comorbidity and complex etiologies present significant challenges for accurate diagnosis and effective treatments. Conventional clinical and experimental studies are time-intensive, burdening research progress considerably. This paper introduces a high-throughput digital twin framework for modeling neurite deteriorations associated with NDDs, integrating synthetic data generation, experimental images, and machine learning (ML) models. The synthetic data generator utilizes an isogeometric analysis (IGA)-based phase field model to capture diverse neurite deterioration patterns such as neurite retraction, atrophy, and fragmentation while mitigating the limitations of scarce experimental data. The ML model utilizes MetaFormer-based gated spatiotemporal attention architecture with deep temporal layers and provides fast predictions. The framework effectively captures long-range temporal dependencies and intricate morphological transformations with average errors of 1.9641% and 6.0339% for synthetic and experimental neurite deterioration, respectively. Seamlessly integrating simulations, experiments, and ML, the digital twin framework can guide researchers to make informed experimental decisions by predicting potential experimental outcomes, significantly reducing costs and saving valuable time. It can also advance our understanding of neurite deterioration and provide a scalable solution for exploring complex neurological mechanisms, contributing to the development of targeted treatments.
Authors: Chengxi Yan, Jie Jian, Yang Li
Abstract: Considering the lack of a unified framework for image description and deep cultural analysis at the subject level in the field of Ancient Chinese Paintings (ACP), this study utilized the Beijing Palace Museum's ACP collections to develop a semantic model integrating the iconological theory with a new workflow for term extraction and mapping. Our findings underscore the model's effectiveness. SDM can be used to support further art-related knowledge organization and cultural exploration of ACPs.
Authors: Meenakshi Krishnan, Liam Fowl, Ramani Duraiswami
Abstract: Differentiable 3D Gaussian splatting has emerged as an efficient and flexible rendering technique for representing complex scenes from a collection of 2D views and enabling high-quality real-time novel-view synthesis. However, its reliance on photometric losses can lead to imprecisely reconstructed geometry and extracted meshes, especially in regions with high curvature or fine detail. We propose a novel regularization method using the gradients of a signed distance function estimated from the Gaussians, to improve the quality of rendering while also extracting a surface mesh. The regularizing normal supervision facilitates better rendering and mesh reconstruction, which is crucial for downstream applications in video generation, animation, AR-VR and gaming. We demonstrate the effectiveness of our approach on datasets such as Mip-NeRF360, Tanks and Temples, and Deep-Blending. Our method scores higher on photorealism metrics compared to other mesh extracting rendering methods without compromising mesh quality.
Authors: Sina Ehsani, Fenglian Pan, Qingpei Hu, Jian Liu
Abstract: Accurate prediction of spatial-temporal (ST) information in dynamic systems, such as urban mobility and weather patterns, is a crucial yet challenging problem. The complexity stems from the intricate interplay between spatial proximity and temporal relevance, where both long-term trends and short-term fluctuations are present in convoluted patterns. Existing approaches, including traditional statistical methods and conventional neural networks, may provide inaccurate results due to the lack of an effective mechanism that simultaneously incorporates information at variable temporal depths while maintaining spatial context, resulting in a trade-off between comprehensive long-term historical analysis and responsiveness to short-term new information. To bridge this gap, this paper proposes the BiDepth Multimodal Neural Network (BDMNN) with bidirectional depth modulation that enables a comprehensive understanding of both long-term seasonality and short-term fluctuations, adapting to the complex ST context. Case studies with real-world public data demonstrate significant improvements in prediction accuracy, with a 12% reduction in Mean Squared Error for urban traffic prediction and a 15% improvement in rain precipitation forecasting compared to state-of-the-art benchmarks, without demanding extra computational resources.
Authors: Juntao Jiang, Jiangning Zhang, Weixuan Liu, Muxuan Gao, Xiaobin Hu, Xiaoxiao Yan, Feiyue Huang, Yong Liu
Abstract: In recent years, there have been significant advancements in deep learning for medical image analysis, especially with convolutional neural networks (CNNs) and transformer models. However, CNNs face limitations in capturing long-range dependencies while transformers suffer high computational complexities. To address this, we propose RWKV-UNet, a novel model that integrates the RWKV (Receptance Weighted Key Value) structure into the U-Net architecture. This integration enhances the model's ability to capture long-range dependencies and improve contextual understanding, which is crucial for accurate medical image segmentation. We build a strong encoder with developed inverted residual RWKV (IR-RWKV) blocks combining CNNs and RWKVs. We also propose a Cross-Channel Mix (CCM) module to improve skip connections with multi-scale feature fusion, achieving global channel information integration. Experiments on benchmark datasets, including Synapse, ACDC, BUSI, CVC-ClinicDB, CVC-ColonDB, Kvasir-SEG, ISIC 2017 and GLAS show that RWKV-UNet achieves state-of-the-art performance on various types of medical image segmentation. Additionally, smaller variants, RWKV-UNet-S and RWKV-UNet-T, balance accuracy and computational efficiency, making them suitable for broader clinical applications.
Authors: Leyla A. Kabuli, Griffin Foster
Abstract: Radar is a low-cost and ubiquitous automotive sensor, but is limited by array resolution and sensitivity when performing direction of arrival analysis. Synthetic Aperture Radar (SAR) is a class of techniques to improve azimuth resolution and sensitivity for radar. Interferometric SAR (InSAR) can be used to extract elevation from the variations in phase measurements in SAR images. Utilizing InSAR we show that a typical, low-resolution radar array mounted on a vehicle can be used to accurately localize detections in 3D space for both urban and agricultural environments. We generate point clouds in each environment by combining InSAR with a signal processing scheme tailored to automotive driving. This low-compute approach allows radar to be used as a primary sensor to map fine details in complex driving environments, and be used to make autonomous perception decisions.
Authors: Kavita Selva, Satita Vittayaareekul, Brando Miranda
Abstract: Currently, data and model size dominate the narrative in the training of super-large, powerful models. However, there has been a lack of exploration on the effect of other attributes of the training dataset on model performance. We hypothesize that dataset diversity can impact the performance of vision models. Our study shows positive correlations between test set accuracy and data diversity, providing an argument for furthering the research of dataset attributes beyond size. We analyzed pre-training and model-agnostic meta-learning methods on twelve popular visual datasets (e.g., Omniglot, CIFAR-FS, Aircraft) and five model configurations, including MAML variants with different numbers of inner gradient steps and supervised learning. We show moderate to strong positive correlations (R-squared: 0.15-0.42) between accuracy and data diversity and weaker but significant correlations (R-squared: ~0.2) between loss and diversity. These findings support our hypothesis and demonstrate a promising way for a deeper exploration of how formal data diversity influences model performance. This initial study highlights the potential of (Task2Vec) data diversity as a valuable measure in the rapidly evolving field of large-scale learning and emphasizes that understanding the dataset is key to building more powerful and generalizable models.
Authors: Donghwi Jung, Keonwoo Kim, Seong-Woo Kim
Abstract: We propose GOTLoc, a robust localization method capable of operating even in outdoor environments where GPS signals are unavailable. The method achieves this robust localization by leveraging comparisons between scene graphs generated from text descriptions and maps. Existing text-based localization studies typically represent maps as point clouds and identify the most similar scenes by comparing embeddings of text and point cloud data. However, point cloud maps have limited scalability as it is impractical to pre-generate maps for all outdoor spaces. Furthermore, their large data size makes it challenging to store and utilize them directly on actual robots. To address these issues, GOTLoc leverages compact data structures, such as scene graphs, to store spatial information, enabling individual robots to carry and utilize large amounts of map data. Additionally, by utilizing publicly available map data, such as OpenStreetMap, which provides global information on outdoor spaces, we eliminate the need for additional effort to create custom map data. For performance evaluation, we utilized the KITTI360Pose dataset in conjunction with corresponding OpenStreetMap data to compare the proposed method with existing approaches. Our results demonstrate that the proposed method achieves accuracy comparable to algorithms relying on point cloud maps. Moreover, in city-scale tests, GOTLoc required significantly less storage compared to point cloud-based methods and completed overall processing within a few seconds, validating its applicability to real-world robotics. Our code is available at https://github.com/donghwijung/GOTLoc.
Authors: Siqi Zhao (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada), Wangyang Li (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada), Xiru Wang (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada), Stevie Foglia (School of Biomedical Engineering, McMaster University, Hamilton, Ontario, Canada), Hongzhao Tan (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada), Bohan Zhang (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada), Ameer Hamoodi (School of Biomedical Engineering, McMaster University, Hamilton, Ontario, Canada), Aimee Nelson (School of Biomedical Engineering, McMaster University, Hamilton, Ontario, Canada, Department of Kinesiology, McMaster University, Hamilton, Ontario, Canada), Zhen Gao (WBooth School of Engineering Practice and Technology, McMaster University, Hamilton, Ontario Canada, School of Biomedical Engineering, McMaster University, Hamilton, Ontario, Canada)
Abstract: Machine learning (ML) and deep learning (DL) techniques have been widely applied to analyze electroencephalography (EEG) signals for disease diagnosis and brain-computer interfaces (BCI). The integration of multimodal data has been shown to enhance the accuracy of ML and DL models. Combining EEG with other modalities can improve clinical decision-making by addressing complex tasks in clinical populations. This systematic literature review explores the use of multimodal EEG data in ML and DL models for clinical applications. A comprehensive search was conducted across PubMed, Web of Science, and Google Scholar, yielding 16 relevant studies after three rounds of filtering. These studies demonstrate the application of multimodal EEG data in addressing clinical challenges, including neuropsychiatric disorders, neurological conditions (e.g., seizure detection), neurodevelopmental disorders (e.g., autism spectrum disorder), and sleep stage classification. Data fusion occurred at three levels: signal, feature, and decision levels. The most commonly used ML models were support vector machines (SVM) and decision trees. Notably, 11 out of the 16 studies reported improvements in model accuracy with multimodal EEG data. This review highlights the potential of multimodal EEG-based ML models in enhancing clinical diagnostics and problem-solving.
Authors: Jiayin Wang, Mingfeng Yao, Yanran Wei, Xiaoyu Guo, Ayong Zheng, Weidong Zhao
Abstract: For Minimally Invasive Surgical (MIS) robots, accurate haptic interaction force feedback is essential for ensuring the safety of interacting with soft tissue. However, most existing MIS robotic systems cannot facilitate direct measurement of the interaction force with hardware sensors due to space limitations. This letter introduces an effective vision-based scheme that utilizes a One-Shot structured light projection with a designed pattern on soft tissue coupled with haptic information processing through a trained image-to-force neural network. The images captured from the endoscopic stereo camera are analyzed to reconstruct high-resolution 3D point clouds for soft tissue deformation. Based on this, a modified PointNet-based force estimation method is proposed, which excels in representing the complex mechanical properties of soft tissue. Numerical force interaction experiments are conducted on three silicon materials with different stiffness. The results validate the effectiveness of the proposed scheme.
Authors: Jussi Kalliola, Lauri Suomela, Sergio Moreschini, David H\"astbacka
Abstract: Localization within a known environment is a crucial capability for mobile robots. Simultaneous Localization and Mapping (SLAM) is a prominent solution to this problem. SLAM is a framework that consists of a diverse set of computational tasks ranging from real-time tracking to computation-intensive map optimization. This combination can present a challenge for resource-limited mobile robots. Previously, edge-assisted SLAM methods have demonstrated promising real-time execution capabilities by offloading heavy computations while performing real-time tracking onboard. However, the common approach of utilizing a client-server architecture for offloading is sensitive to server and network failures. In this article, we propose a novel edge-assisted SLAM framework capable of self-organizing fully distributed SLAM execution across a network of devices or functioning on a single device without connectivity. The architecture consists of three layers and is designed to be device-agnostic, resilient to network failures, and minimally invasive to the core SLAM system. We have implemented and demonstrated the framework for monocular ORB SLAM3 and evaluated it in both fully distributed and standalone SLAM configurations against the ORB SLAM3. The experiment results demonstrate that the proposed design matches the accuracy and resource utilization of the monolithic approach while enabling collaborative execution.
Authors: Laurenz Nagler, Martin Zach, Thomas Pock
Abstract: Diffusion models have recently shown remarkable results in magnetic resonance imaging reconstruction. However, the employed networks typically are black-box estimators of the (smoothed) prior score with tens of millions of parameters, restricting interpretability and increasing reconstruction time. Furthermore, parallel imaging reconstruction algorithms either rely on off-line coil sensitivity estimation, which is prone to misalignment and restricting sampling trajectories, or perform per-coil reconstruction, making the computational cost proportional to the number of coils. To overcome this, we jointly reconstruct the image and the coil sensitivities using the lightweight, parameter-efficient, and interpretable product of Gaussian mixture diffusion model as an image prior and a classical smoothness priors on the coil sensitivities. The proposed method delivers promising results while allowing for fast inference and demonstrating robustness to contrast out-of-distribution data and sampling trajectories, comparable to classical variational penalties such as total variation. Finally, the probabilistic formulation allows the calculation of the posterior expectation and pixel-wise variance.
Authors: Bailiang Jian, Jiazhen Pan, Yitong Li, Fabian Bongratz, Ruochen Li, Daniel Rueckert, Benedikt Wiestler, Christian Wachinger
Abstract: Predicting future brain states is crucial for understanding healthy aging and neurodegenerative diseases. Longitudinal brain MRI registration, a cornerstone for such analyses, has long been limited by its inability to forecast future developments, reliance on extensive, dense longitudinal data, and the need to balance registration accuracy with temporal smoothness. In this work, we present \emph{TimeFlow}, a novel framework for longitudinal brain MRI registration that overcomes all these challenges. Leveraging a U-Net architecture with temporal conditioning inspired by diffusion models, TimeFlow enables accurate longitudinal registration and facilitates prospective analyses through future image prediction. Unlike traditional methods that depend on explicit smoothness regularizers and dense sequential data, TimeFlow achieves temporal consistency and continuity without these constraints. Experimental results highlight its superior performance in both future timepoint prediction and registration accuracy compared to state-of-the-art methods. Additionally, TimeFlow supports novel biological brain aging analyses, effectively differentiating neurodegenerative conditions from healthy aging. It eliminates the need for segmentation, thereby avoiding the challenges of non-trivial annotation and inconsistent segmentation errors. TimeFlow paves the way for accurate, data-efficient, and annotation-free prospective analyses of brain aging and chronic diseases.
Authors: Sheng Hong, Chunran Zheng, Yishu Shen, Changze Li, Fu Zhang, Tong Qin, Shaojie Shen
Abstract: In recent years, 3D Gaussian splatting (3D-GS) has emerged as a novel scene representation approach. However, existing vision-only 3D-GS methods often rely on hand-crafted heuristics for point-cloud densification and face challenges in handling occlusions and high GPU memory and computation consumption. LiDAR-Inertial-Visual (LIV) sensor configuration has demonstrated superior performance in localization and dense mapping by leveraging complementary sensing characteristics: rich texture information from cameras, precise geometric measurements from LiDAR, and high-frequency motion data from IMU. Inspired by this, we propose a novel real-time Gaussian-based simultaneous localization and mapping (SLAM) system. Our map system comprises a global Gaussian map and a sliding window of Gaussians, along with an IESKF-based odometry. The global Gaussian map consists of hash-indexed voxels organized in a recursive octree, effectively covering sparse spatial volumes while adapting to different levels of detail and scales. The Gaussian map is initialized through multi-sensor fusion and optimized with photometric gradients. Our system incrementally maintains a sliding window of Gaussians, significantly reducing GPU computation and memory consumption by only optimizing the map within the sliding window. Moreover, we implement a tightly coupled multi-sensor fusion odometry with an iterative error state Kalman filter (IESKF), leveraging real-time updating and rendering of the Gaussian map. Our system represents the first real-time Gaussian-based SLAM framework deployable on resource-constrained embedded systems, demonstrated on the NVIDIA Jetson Orin NX platform. The framework achieves real-time performance while maintaining robust multi-sensor fusion capabilities. All implementation algorithms, hardware designs, and CAD models will be publicly available.
Authors: Tianru Zhang, Li Ju, Prashant Singh, Salman Toor
Abstract: Analyzing large-scale datasets, especially involving complex and high-dimensional data like images, is particularly challenging. While self-supervised learning (SSL) has proven effective for learning representations from unlabelled data, it typically focuses on flat, non-hierarchical structures, missing the multi-level relationships present in many real-world datasets. Hierarchical clustering (HC) can uncover these relationships by organizing data into a tree-like structure, but it often relies on rigid similarity metrics that struggle to capture the complexity of diverse data types. To address these we envision $\texttt{InfoHier}$, a framework that combines SSL with HC to jointly learn robust latent representations and hierarchical structures. This approach leverages SSL to provide adaptive representations, enhancing HC's ability to capture complex patterns. Simultaneously, it integrates HC loss to refine SSL training, resulting in representations that are more attuned to the underlying information hierarchy. $\texttt{InfoHier}$ has the potential to improve the expressiveness and performance of both clustering and representation learning, offering significant benefits for data analysis, management, and information retrieval.
Authors: Shao-Hao Lu, Ren Wang, Ching-Chun Huang, Wei-Chen Chiu
Abstract: Recently, diffusion-based blind super-resolution (SR) methods have shown great ability to generate high-resolution images with abundant high-frequency detail, but the detail is often achieved at the expense of fidelity. Meanwhile, another line of research focusing on rectifying the reverse process of diffusion models (i.e., diffusion guidance), has demonstrated the power to generate high-fidelity results for non-blind SR. However, these methods rely on known degradation kernels, making them difficult to apply to blind SR. To address these issues, we introduce degradation-aware models that can be integrated into the diffusion guidance framework, eliminating the need to know degradation kernels. Additionally, we propose two novel techniques input perturbation and guidance scalar to further improve our performance. Extensive experimental results show that our proposed method has superior performance over state-of-the-art methods on blind SR benchmarks
Authors: Kuicai Dong, Yujing Chang, Xin Deik Goh, Dexun Li, Ruiming Tang, Yong Liu
Abstract: Multi-modal document retrieval is designed to identify and retrieve various forms of multi-modal content, such as figures, tables, charts, and layout information from extensive documents. Despite its significance, there is a notable lack of a robust benchmark to effectively evaluate the performance of systems in multi-modal document retrieval. To address this gap, this work introduces a new benchmark, named as MMDocIR, encompassing two distinct tasks: page-level and layout-level retrieval. The former focuses on localizing the most relevant pages within a long document, while the latter targets the detection of specific layouts, offering a more fine-grained granularity than whole-page analysis. A layout can refer to a variety of elements such as textual paragraphs, equations, figures, tables, or charts. The MMDocIR benchmark comprises a rich dataset featuring expertly annotated labels for 1,685 questions and bootstrapped labels for 173,843 questions, making it a pivotal resource for advancing multi-modal document retrieval for both training and evaluation. Through rigorous experiments, we reveal that (i) visual retrievers significantly outperform their text counterparts, (ii) MMDocIR train set can effectively benefit the training process of multi-modal document retrieval and (iii) text retrievers leveraging on VLM-text perform much better than those using OCR-text. These findings underscores the potential advantages of integrating visual elements for multi-modal document retrieval.
Authors: Yan Zhu, Huan Ma, Changqing Zhang
Abstract: With the development of Vision Foundation Models (VFMs) in recent years, Visual In-Context Learning (VICL) has become a better choice compared to modifying models in most scenarios. Different from retraining or fine-tuning model, VICL does not require modifications to the model's weights or architecture, and only needs a prompt with demonstrations to teach VFM how to solve tasks. Currently, significant computational cost for finding optimal prompts for every test sample hinders the deployment of VICL, as determining which demonstrations to use for constructing prompts is very costly. In this paper, however, we find a counterintuitive phenomenon that most test samples actually achieve optimal performance under the same prompts, and searching for sample-level prompts only costs more time but results in completely identical prompts. Therefore, we propose task-level prompting to reduce the cost of searching for prompts during the inference stage and introduce two time-saving yet effective task-level prompt search strategies. Extensive experimental results show that our proposed method can identify near-optimal prompts and reach the best VICL performance with a minimal cost that prior work has never achieved.
Authors: Sneha N. Naik, Elsa D. Angelini, Eric A. Hoffman, Elizabeth C. Oelsner, R. Graham Barr, Benjamin M. Smith, Andrew F. Laine
Abstract: The ratio of airway tree lumen to lung size (ALR), assessed at full inspiration on high resolution full-lung computed tomography (CT), is a major risk factor for chronic obstructive pulmonary disease (COPD). There is growing interest to infer ALR from cardiac CT images, which are widely available in epidemiological cohorts, to investigate the relationship of ALR to severe COVID-19 and post-acute sequelae of SARS-CoV-2 infection (PASC). Previously, cardiac scans included approximately 2/3 of the total lung volume with 5-6x greater slice thickness than high-resolution (HR) full-lung (FL) CT. In this study, we present a novel attention-based Multi-view Swin Transformer to infer FL ALR values from segmented cardiac CT scans. For the supervised training we exploit paired full-lung and cardiac CTs acquired in the Multi-Ethnic Study of Atherosclerosis (MESA). Our network significantly outperforms a proxy direct ALR inference on segmented cardiac CT scans and achieves accuracy and reproducibility comparable with a scan-rescan reproducibility of the FL ALR ground-truth.
Authors: Suraj Pai (Artificial Intelligence in Medicine, Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Ibrahim Hadzic (Artificial Intelligence in Medicine, Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Dennis Bontempi (Artificial Intelligence in Medicine, Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Keno Bressem (Department of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine and Health, Klinikum rechts der Isar, TUM University Hospital, Department of Cardiovascular Radiology and Nuclear Medicine, Technical University of Munich, School of Medicine and Health, German Heart Center, TUM University Hospital), Benjamin H. Kann (Artificial Intelligence in Medicine, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Andriy Fedorov (Department of Radiology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Raymond H. Mak (Artificial Intelligence in Medicine, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School), Hugo J. W. L. Aerts (Artificial Intelligence in Medicine, Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School, Department of Radiology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School)
Abstract: Foundation models (FMs) have shown transformative potential in radiology by performing diverse, complex tasks across imaging modalities. Here, we developed CT-FM, a large-scale 3D image-based pre-trained model designed explicitly for various radiological tasks. CT-FM was pre-trained using 148,000 computed tomography (CT) scans from the Imaging Data Commons through label-agnostic contrastive learning. We evaluated CT-FM across four categories of tasks, namely, whole-body and tumor segmentation, head CT triage, medical image retrieval, and semantic understanding, showing superior performance against state-of-the-art models. Beyond quantitative success, CT-FM demonstrated the ability to cluster regions anatomically and identify similar anatomical and structural concepts across scans. Furthermore, it remained robust across test-retest settings and indicated reasonable salient regions attached to its embeddings. This study demonstrates the value of large-scale medical imaging foundation models and by open-sourcing the model weights, code, and data, aims to support more adaptable, reliable, and interpretable AI solutions in radiology.
Authors: Xingchi Chen, Xiuyi Jia, Zhuoran Zheng
Abstract: Currently, transformer-based algorithms are making a splash in the domain of image deblurring. Their achievement depends on the self-attention mechanism with CNN stem to model long range dependencies between tokens. Unfortunately, this ear-pleasing pipeline introduces high computational complexity and makes it difficult to run an ultra-high-definition image on a single GPU in real time. To trade-off accuracy and efficiency, the input degraded image is computed cyclically over three dimensional ($C$, $W$, and $H$) signals without a self-attention mechanism. We term this deep network as Multi-scale Cubic-Mixer, which is acted on both the real and imaginary components after fast Fourier transform to estimate the Fourier coefficients and thus obtain a deblurred image. Furthermore, we combine the multi-scale cubic-mixer with a slicing strategy to generate high-quality results at a much lower computational cost. Experimental results demonstrate that the proposed algorithm performs favorably against the state-of-the-art deblurring approaches on the several benchmarks and a new ultra-high-definition dataset in terms of accuracy and speed.
Authors: Songwei Ge, Taesung Park, Jun-Yan Zhu, Jia-Bin Huang
Abstract: Plain text has become a prevalent interface for text-to-image synthesis. However, its limited customization options hinder users from accurately describing desired outputs. For example, plain text makes it hard to specify continuous quantities, such as the precise RGB color value or importance of each word. Furthermore, creating detailed text prompts for complex scenes is tedious for humans to write and challenging for text encoders to interpret. To address these challenges, we propose using a rich-text editor supporting formats such as font style, size, color, and footnote. We extract each word's attributes from rich text to enable local style control, explicit token reweighting, precise color rendering, and detailed region synthesis. We achieve these capabilities through a region-based diffusion process. We first obtain each word's region based on attention maps of a diffusion process using plain text. For each region, we enforce its text attributes by creating region-specific detailed prompts and applying region-specific guidance, and maintain its fidelity against plain-text generation through region-based injections. We present various examples of image generation from rich text and demonstrate that our method outperforms strong baselines with quantitative evaluations.
Authors: Chih-Hao Lin, Bohan Liu, Yi-Ting Chen, Kuan-Sheng Chen, David Forsyth, Jia-Bin Huang, Anand Bhattad, Shenlong Wang
Abstract: We present UrbanIR (Urban Scene Inverse Rendering), a new inverse graphics model that enables realistic, free-viewpoint renderings of scenes under various lighting conditions with a single video. It accurately infers shape, albedo, visibility, and sun and sky illumination from wide-baseline videos, such as those from car-mounted cameras, differing from NeRF's dense view settings. In this context, standard methods often yield subpar geometry and material estimates, such as inaccurate roof representations and numerous 'floaters'. UrbanIR addresses these issues with novel losses that reduce errors in inverse graphics inference and rendering artifacts. Its techniques allow for precise shadow volume estimation in the original scene. The model's outputs support controllable editing, enabling photorealistic free-viewpoint renderings of night simulations, relit scenes, and inserted objects, marking a significant improvement over existing state-of-the-art methods.
Authors: Julio Silva-Rodr\'iguez, Hadi Chakor, Riadh Kobbi, Jose Dolz, Ismail Ben Ayed
Abstract: Foundation vision-language models are currently transforming computer vision, and are on the rise in medical imaging fueled by their very promising generalization capabilities. However, the initial attempts to transfer this new paradigm to medical imaging have shown less impressive performances than those observed in other domains, due to the significant domain shift and the complex, expert domain knowledge inherent to medical-imaging tasks. Motivated by the need for domain-expert foundation models, we present FLAIR, a pre-trained vision-language model for universal retinal fundus image understanding. To this end, we compiled 38 open-access, mostly categorical fundus imaging datasets from various sources, with up to 101 different target conditions and 288,307 images. We integrate the expert's domain knowledge in the form of descriptive textual prompts, during both pre-training and zero-shot inference, enhancing the less-informative categorical supervision of the data. Such a textual expert's knowledge, which we compiled from the relevant clinical literature and community standards, describes the fine-grained features of the pathologies as well as the hierarchies and dependencies between them. We report comprehensive evaluations, which illustrate the benefit of integrating expert knowledge and the strong generalization capabilities of FLAIR under difficult scenarios with domain shifts or unseen categories. When adapted with a lightweight linear probe, FLAIR outperforms fully-trained, dataset-focused models, more so in the few-shot regimes. Interestingly, FLAIR outperforms by a wide margin larger-scale generalist image-language models and retina domain-specific self-supervised networks, which emphasizes the potential of embedding experts' domain knowledge and the limitations of generalist models in medical imaging.
Authors: Joseph Paul Cohen, Louis Blankemeier, Akshay Chaudhari
Abstract: Models driven by spurious correlations often yield poor generalization performance. We propose the counterfactual (CF) alignment method to detect and quantify spurious correlations of black box classifiers. Our methodology is based on counterfactual images generated with respect to one classifier being input into other classifiers to see if they also induce changes in the outputs of these classifiers. The relationship between these responses can be quantified and used to identify specific instances where a spurious correlation exists. This is validated by observing intuitive trends in face-attribute and waterbird classifiers, as well as by fabricating spurious correlations and detecting their presence, both visually and quantitatively. Furthermore, utilizing the CF alignment method, we demonstrate that we can evaluate robust optimization methods (GroupDRO, JTT, and FLAC) by detecting a reduction in spurious correlations.
Authors: Nir Yellinek, Leonid Karlinsky, Raja Giryes
Abstract: Vision-Language models (VLMs) have proven to be effective at aligning image and text representations, producing superior zero-shot results when transferred to many downstream tasks. However, these representations suffer from some key shortcomings in understanding Compositional Language Concepts (CLC), such as recognizing objects' attributes, states, and relations between different objects. Moreover, VLMs typically have poor interpretability, making it challenging to debug and mitigate compositional-understanding failures. In this work, we introduce the architecture and training technique of Tree-augmented Vision-Language (3VL) model accompanied by our proposed Anchor inference method and Differential Relevance (DiRe) interpretability tool. By expanding the text of an arbitrary image-text pair into a hierarchical tree structure using language analysis tools, 3VL allows the induction of this structure into the visual representation learned by the model, enhancing its interpretability and compositional reasoning. Additionally, we show how Anchor, a simple technique for text unification, can be used to filter nuisance factors while increasing CLC understanding performance, e.g., on the fundamental VL-Checklist benchmark. We also show how DiRe, which performs a differential comparison between VLM relevancy maps, enables us to generate compelling visualizations of the reasons for a model's success or failure. Our code is available at: https://github.com/niryellinek/3VL.
Authors: Haopeng Li, Andong Deng, Jun Liu, Hossein Rahmani, Yulan Guo, Bernt Schiele, Mohammed Bennamoun, Qiuhong Ke
Abstract: Reasoning over sports videos for question answering is an important task with numerous applications, such as player training and information retrieval. However, this task has not been explored due to the lack of relevant datasets and the challenging nature it presents. Most datasets for video question answering (VideoQA) focus mainly on general and coarse-grained understanding of daily-life videos, which is not applicable to sports scenarios requiring professional action understanding and fine-grained motion analysis. In this paper, we introduce the first dataset, named Sports-QA, specifically designed for the sports VideoQA task. The Sports-QA dataset includes various types of questions, such as descriptions, chronologies, causalities, and counterfactual conditions, covering multiple sports. Furthermore, to address the characteristics of the sports VideoQA task, we propose a new Auto-Focus Transformer (AFT) capable of automatically focusing on particular scales of temporal information for question answering. We conduct extensive experiments on Sports-QA, including baseline studies and the evaluation of different methods. The results demonstrate that our AFT achieves state-of-the-art performance.
Authors: Jiahe Chen, Jinkun Cao, Dahua Lin, Kris Kitani, Jiangmiao Pang
Abstract: To predict future trajectories, the normalizing flow with a standard Gaussian prior suffers from weak diversity. The ineffectiveness comes from the conflict between the fact of asymmetric and multi-modal distribution of likely outcomes and symmetric and single-modal original distribution and supervision losses. Instead, we propose constructing a mixed Gaussian prior for a normalizing flow model for trajectory prediction. The prior is constructed by analyzing the trajectory patterns in the training samples without requiring extra annotations while showing better expressiveness and being multi-modal and asymmetric. Besides diversity, it also provides better controllability for probabilistic trajectory generation. We name our method Mixed Gaussian Flow (MGF). It achieves state-of-the-art performance in the evaluation of both trajectory alignment and diversity on the popular UCY/ETH and SDD datasets. Code is available at https://github.com/mulplue/MGF.
Authors: Weixia Zhang, Dingquan Li, Guangtao Zhai, Xiaokang Yang, Kede Ma
Abstract: Contemporary no-reference image quality assessment (NR-IQA) models can effectively quantify perceived image quality, often achieving strong correlations with human perceptual scores on standard IQA benchmarks. Yet, limited efforts have been devoted to treating NR-IQA models as natural image priors for real-world image enhancement, and consequently comparing them from a perceptual optimization standpoint. In this work, we show -- for the first time -- that NR-IQA models can be plugged into the maximum a posteriori (MAP) estimation framework for image enhancement. This is achieved by performing gradient ascent in the diffusion latent space rather than in the raw pixel domain, leveraging a pretrained differentiable and bijective diffusion process. Likely, different NR-IQA models lead to different enhanced outputs, which in turn provides a new computational means of comparing them. Unlike conventional correlation-based measures, our comparison method offers complementary insights into the respective strengths and weaknesses of the competing NR-IQA models in perceptual optimization scenarios. Additionally, we aim to improve the best-performing NR-IQA model in diffusion latent MAP estimation by incorporating the advantages of other top-performing methods. The resulting model delivers noticeably better results in enhancing real-world images afflicted by unknown and complex distortions, all preserving a high degree of image fidelity.
Authors: Hanzhou Liu, Binghan Li, Chengkai Liu, Mi Lu
Abstract: Recent deblurring networks have effectively restored clear images from the blurred ones. However, they often struggle with generalization to unknown domains. Moreover, these models typically focus on distortion metrics such as PSNR and SSIM, neglecting the critical aspect of metrics aligned with human perception. To address these limitations, we propose DeblurDiNAT, a deblurring Transformer based on Dilated Neighborhood Attention. First, DeblurDiNAT employs an alternating dilation factor paradigm to capture both local and global blurred patterns, enhancing generalization and perceptual clarity. Second, a local cross-channel learner aids the Transformer block to understand the short-range relationships between adjacent channels. Additionally, we present a linear feed-forward network with a simple while effective design. Finally, a dual-stage feature fusion module is introduced as an alternative to the existing approach, which efficiently process multi-scale visual information across network levels. Compared to state-of-the-art models, our compact DeblurDiNAT demonstrates superior generalization capabilities and achieves remarkable performance in perceptual metrics, while maintaining a favorable model size.
Authors: Qing Wu, Xu Guo, Lixuan Chen, Yanyan Liu, Dongming He, Xudong Wang, Xueli Chen, Yifeng Zhang, S. Kevin Zhou, Jingyi Yu, Yuyao Zhang
Abstract: X-ray CT often suffers from shadowing and streaking artifacts in the presence of metallic materials, which severely degrade imaging quality. Physically, the linear attenuation coefficients (LACs) of metals vary significantly with X-ray energy, causing a nonlinear beam hardening effect (BHE) in CT measurements. Reconstructing CT images from metal-corrupted measurements consequently becomes a challenging nonlinear inverse problem. Existing state-of-the-art (SOTA) metal artifact reduction (MAR) algorithms rely on supervised learning with numerous paired CT samples. While promising, these supervised methods often assume that the unknown LACs are energy-independent, ignoring the energy-induced BHE, which results in limited generalization. Moreover, the requirement for large datasets also limits their applications in real-world scenarios. In this work, we propose Density neural representation (Diner), a novel unsupervised MAR method. Our key innovation lies in formulating MAR as an energy-independent density reconstruction problem that strictly adheres to the photon-tissue absorption physical model. This model is inherently nonlinear and complex, making it a rarely considered approach in inverse imaging problems. By introducing the water-equivalent tissues approximation and a new polychromatic model to characterize the nonlinear CT acquisition process, we directly learn the neural representation of the density map from raw measurements without using external training data. This energy-independent density reconstruction framework fundamentally resolves the nonlinear BHE, enabling superior MAR performance across a wide range of scanning scenarios. Extensive experiments on both simulated and real-world datasets demonstrate the superiority of our unsupervised Diner over popular supervised methods in terms of MAR performance and robustness.
Authors: Cong Cao, Huanjing Yue, Xin Liu, Jingyu Yang
Abstract: Diffusion-based zero-shot image restoration and enhancement models have achieved great success in various tasks of image restoration and enhancement. However, directly applying them to video restoration and enhancement results in severe temporal flickering artifacts. In this paper, we propose the first framework for zero-shot video restoration and enhancement based on the pre-trained image diffusion model. By replacing the spatial self-attention layer with the proposed short-long-range (SLR) temporal attention layer, the pre-trained image diffusion model can take advantage of the temporal correlation between frames. We further propose temporal consistency guidance, spatial-temporal noise sharing, and an early stopping sampling strategy to improve temporally consistent sampling. Our method is a plug-and-play module that can be inserted into any diffusion-based image restoration or enhancement methods to further improve their performance. Experimental results demonstrate the superiority of our proposed method. Our code is available at https://github.com/cao-cong/ZVRD.
Authors: Dominik Winter, Nicolas Triltsch, Marco Rosati, Anatoliy Shumilov, Ziya Kokaragac, Yuri Popov, Thomas Padel, Laura Sebastian Monasor, Ross Hill, Markus Schick, Nicolas Brieu
Abstract: Creating in-silico data with generative AI promises a cost-effective alternative to staining, imaging, and annotating whole slide images in computational pathology. Diffusion models are the state-of-the-art solution for generating in-silico images, offering unparalleled fidelity and realism. Using appearance transfer diffusion models allows for zero-shot image generation, facilitating fast application and making model training unnecessary. However current appearance transfer diffusion models are designed for natural images, where the main task is to transfer the foreground object from an origin to a target domain, while the background is of insignificant importance. In computational pathology, specifically in oncology, it is however not straightforward to define which objects in an image should be classified as foreground and background, as all objects in an image may be of critical importance for the detailed understanding the tumor micro-environment. We contribute to the applicability of appearance transfer diffusion models to immunohistochemistry-stained images by modifying the appearance transfer guidance to alternate between class-specific AdaIN feature statistics matchings using existing segmentation masks. The performance of the proposed method is demonstrated on the downstream task of supervised epithelium segmentation, showing that the number of manual annotations required for model training can be reduced by 75%, outperforming the baseline approach. Additionally, we consulted with a certified pathologist to investigate future improvements. We anticipate this work to inspire the application of zero-shot diffusion models in computational pathology, providing an efficient method to generate in-silico images with unmatched fidelity and realism, which prove meaningful for downstream tasks, such as training existing deep learning models or finetuning foundation models.
Authors: Kaiyue Sun, Kaiyi Huang, Xian Liu, Yue Wu, Zihan Xu, Zhenguo Li, Xihui Liu
Abstract: Text-to-video (T2V) generative models have advanced significantly, yet their ability to compose different objects, attributes, actions, and motions into a video remains unexplored. Previous text-to-video benchmarks also neglect this important ability for evaluation. In this work, we conduct the first systematic study on compositional text-to-video generation. We propose T2V-CompBench, the first benchmark tailored for compositional text-to-video generation. T2V-CompBench encompasses diverse aspects of compositionality, including consistent attribute binding, dynamic attribute binding, spatial relationships, motion binding, action binding, object interactions, and generative numeracy. We further carefully design evaluation metrics of multimodal large language model (MLLM)-based, detection-based, and tracking-based metrics, which can better reflect the compositional text-to-video generation quality of seven proposed categories with 1400 text prompts. The effectiveness of the proposed metrics is verified by correlation with human evaluations. We also benchmark various text-to-video generative models and conduct in-depth analysis across different models and various compositional categories. We find that compositional text-to-video generation is highly challenging for current models, and we hope our attempt could shed light on future research in this direction.
Authors: Isaac Osei Agyemang, Isaac Adjei-Mensah, Daniel Acheampong, Gordon Owusu Boateng, Adu Asare Baffour
Abstract: Structural health monitoring (SHM) is essential for ensuring the safety and longevity of infrastructure, but complex image environments, noisy labels, and reliance on manual damage assessments often hinder its effectiveness. This study introduces the Guided Detection Network (Guided-DetNet), a framework designed to address these challenges. Guided-DetNet is characterized by a Generative Attention Module (GAM), Hierarchical Elimination Algorithm (HEA), and Volumetric Contour Visual Assessment (VCVA). GAM leverages cross-horizontal and cross-vertical patch merging and cross-foreground-background feature fusion to generate varied features to mitigate complex image environments. HEA addresses noisy labeling using hierarchical relationships among classes to refine instances given an image by eliminating unlikely class instances. VCVA assesses the severity of detected damages via volumetric representation and quantification leveraging the Dirac delta distribution. A comprehensive quantitative study and two robustness tests were conducted using the PEER Hub dataset, and a drone-based application, which involved a field experiment, was conducted to substantiate Guided-DetNet's promising performances. In triple classification tasks, the framework achieved 96% accuracy, surpassing state-of-the-art classifiers by up to 3%. In dual detection tasks, it outperformed competitive detectors with a precision of 94% and a mean average precision (mAP) of 79% while maintaining a frame rate of 57.04fps, suitable for real-time applications. Additionally, robustness tests demonstrated resilience under adverse conditions, with precision scores ranging from 79% to 91%. Guided-DetNet is established as a robust and efficient framework for SHM, offering advancements in automation and precision, with the potential for widespread application in drone-based infrastructure inspections.
Authors: Zijian Zhao, Tingwei Chen, Zhijie Cai, Xiaoyang Li, Hang Li, Qimei Chen, Guangxu Zhu
Abstract: In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available at https://github.com/RS2002/CrossFi.
Authors: Qiang Zheng, Chao Zhang, Jian Sun
Abstract: Point cloud classification plays a crucial role in the processing and analysis of data from 3D sensors such as LiDAR, which are commonly used in applications like autonomous vehicles, robotics, and environmental monitoring. However, traditional neural networks, which rely heavily on multiplication operations, often face challenges in terms of high computational costs and energy consumption. This study presents a novel family of efficient MLP-based architectures designed to improve the computational efficiency of point cloud classification tasks in sensor systems. The baseline model, Mul-MLP, utilizes conventional multiplication operations, while Add-MLP and Shift-MLP replace multiplications with addition and shift operations, respectively. These replacements leverage more sensor-friendly operations that can significantly reduce computational overhead, making them particularly suitable for resource-constrained sensor platforms. To further enhance performance, we propose SA-MLP, a hybrid architecture that alternates between shift and adder layers, preserving the network depth while optimizing computational efficiency. Unlike previous approaches such as ShiftAddNet, which increase the layer count and limit representational capacity by freezing shift weights, SA-MLP fully exploits the complementary advantages of shift and adder layers by employing distinct learning rates and optimizers. Experimental results show that Add-MLP and Shift-MLP achieve competitive performance compared to Mul-MLP, while SA-MLP surpasses the baseline, delivering results comparable to state-of-the-art MLP models in terms of both classification accuracy and computational efficiency. This work offers a promising, energy-efficient solution for sensor-driven applications requiring real-time point cloud classification, particularly in environments with limited computational resources.
Authors: Chen Hu, Yian Huang, Kexuan Li, Luping Zhang, Chang Long, Yiming Zhu, Tian Pu, Zhenming Peng
Abstract: Infrared small target detection (ISTD) is widely used in civilian and military applications. However, ISTD encounters several challenges, including the tendency for small and dim targets to be obscured by complex backgrounds.To address this issue, we propose the Dynamic Attention Transformer Network (DATransNet), which aims to extract and preserve edge information of small targets.DATransNet employs the Dynamic Attention Transformer (DATrans), simulating central difference convolutions (CDC) to extract and integrate gradient features with deeper features.Furthermore, we propose a global feature extraction module (GFEM) that offers a comprehensive perspective to prevent the network from focusing solely on details while neglecting the background information. We compare the network with state-of-the-art (SOTA) approaches, and the results demonstrate that our method performs effectively. Our source code is available at https://github.com/greekinRoma/DATransNet.
Authors: Kevin Tirta Wijaya, Christofel Rio Goenawan, Seung-Hyun Kong
Abstract: Point cloud completion networks are conventionally trained to minimize the disparities between the completed point cloud and the ground-truth counterpart. However, an incomplete object-level point cloud can have multiple valid completion solutions when it is examined in isolation. This one-to-many mapping issue can cause contradictory supervision signals to the network because the loss function may produce different values for identical input-output pairs of the network. In many cases, this issue could adversely affect the network optimization process. In this work, we propose to enhance the conventional learning objective using a novel completion consistency loss to mitigate the one-to-many mapping problem. Specifically, the proposed consistency loss ensure that a point cloud completion network generates a coherent completion solution for incomplete objects originating from the same source point cloud. Experimental results across multiple well-established datasets and benchmarks demonstrated the proposed completion consistency loss have excellent capability to enhance the completion performance of various existing networks without any modification to the design of the networks. The proposed consistency loss enhances the performance of the point completion network without affecting the inference speed, thereby increasing the accuracy of point cloud completion. Notably, a state-of-the-art point completion network trained with the proposed consistency loss can achieve state-of-the-art accuracy on the challenging new MVP dataset. The code and result of experiment various point completion models using proposed consistency loss will be available at: https://github.com/kaist-avelab/ConsistencyLoss .
Authors: Shengxun Wei, Zan Gao, Chunjie Ma, Yibo Zhao, Weili Guan, Shengyong Chen
Abstract: Cloth-changing person re-identification is a subject closer to the real world, which focuses on solving the problem of person re-identification after pedestrians change clothes. The primary challenge in this field is to overcome the complex interplay between intra-class and inter-class variations and to identify features that remain unaffected by changes in appearance. Sufficient data collection for model training would significantly aid in addressing this problem. However, it is challenging to gather diverse datasets in practice. Current methods focus on implicitly learning identity information from the original image or introducing additional auxiliary models, which are largely limited by the quality of the image and the performance of the additional model. To address these issues, inspired by prompt learning, we propose a novel multiple information prompt learning (MIPL) scheme for cloth-changing person ReID, which learns identity robust features through the common prompt guidance of multiple messages. Specifically, the clothing information stripping (CIS) module is designed to decouple the clothing information from the original RGB image features to counteract the influence of clothing appearance. The Bio-guided attention (BGA) module is proposed to increase the learning intensity of the model for key information. A dual-length hybrid patch (DHP) module is employed to make the features have diverse coverage to minimize the impact of feature bias. Extensive experiments demonstrate that the proposed method outperforms all state-of-the-art methods on the LTCC, Celeb-reID, Celeb-reID-light, and CSCC datasets, achieving rank-1 scores of 74.8%, 73.3%, 66.0%, and 88.1%, respectively. When compared to AIM (CVPR23), ACID (TIP23), and SCNet (MM23), MIPL achieves rank-1 improvements of 11.3%, 13.8%, and 7.9%, respectively, on the PRCC dataset.
Authors: Huiwon Jang, Sihyun Yu, Jinwoo Shin, Pieter Abbeel, Younggyo Seo
Abstract: Efficient tokenization of videos remains a challenge in training vision models that can process long videos. One promising direction is to develop a tokenizer that can encode long video clips, as it would enable the tokenizer to leverage the temporal coherence of videos better for tokenization. However, training existing tokenizers on long videos often incurs a huge training cost as they are trained to reconstruct all the frames at once. In this paper, we introduce CoordTok, a video tokenizer that learns a mapping from coordinate-based representations to the corresponding patches of input videos, inspired by recent advances in 3D generative models. In particular, CoordTok encodes a video into factorized triplane representations and reconstructs patches that correspond to randomly sampled $(x,y,t)$ coordinates. This allows for training large tokenizer models directly on long videos without requiring excessive training resources. Our experiments show that CoordTok can drastically reduce the number of tokens for encoding long video clips. For instance, CoordTok can encode a 128-frame video with 128$\times$128 resolution into 1280 tokens, while baselines need 6144 or 8192 tokens to achieve similar reconstruction quality. We further show that this efficient video tokenization enables memory-efficient training of a diffusion transformer that can generate 128 frames at once.
Authors: Zhenxiong Tan, Songhua Liu, Xingyi Yang, Qiaochu Xue, Xinchao Wang
Abstract: In this paper, we introduce OminiControl, a highly versatile and parameter-efficient framework that integrates image conditions into pre-trained Diffusion Transformer (DiT) models. At its core, OminiControl leverages a parameter reuse mechanism, enabling the DiT to encode image conditions using itself as a powerful backbone and process them with its flexible multi-modal attention processors. Unlike existing methods, which rely heavily on additional encoder modules with complex architectures, OminiControl (1) effectively and efficiently incorporates injected image conditions with only ~0.1% additional parameters, and (2) addresses a wide range of image conditioning tasks in a unified manner, including subject-driven generation and spatially-aligned conditions such as edges, depth, and more. Remarkably, these capabilities are achieved by training on images generated by the DiT itself, which is particularly beneficial for subject-driven generation. Extensive evaluations demonstrate that OminiControl outperforms existing UNet-based and DiT-adapted models in both subject-driven and spatially-aligned conditional generation. Additionally, we release our training dataset, Subjects200K, a diverse collection of over 200,000 identity-consistent images, along with an efficient data synthesis pipeline to advance research in subject-consistent generation.
Authors: Tingxu Han, Weisong Sun, Yanrong Hu, Chunrong Fang, Yonglong Zhang, Shiqing Ma, Tao Zheng, Zhenyu Chen, Zhenting Wang
Abstract: Text-to-image diffusion models have shown an impressive ability to generate high-quality images from input textual descriptions. However, concerns have been raised about the potential for these models to create content that infringes on copyrights or depicts disturbing subject matter. Removing specific concepts from these models is a promising potential solution to this problem. However, existing methods for concept removal do not work well in practical but challenging scenarios where concepts need to be continuously removed. Specifically, these methods lead to poor alignment between the text prompts and the generated image after the continuous removal process. To address this issue, we propose a novel approach called CCRT that includes a designed knowledge distillation paradigm. It constrains the text-image alignment behavior during the continuous concept removal process by using a set of text prompts generated through our genetic algorithm, which employs a designed fuzzing strategy. We conduct extensive experiments involving the removal of various concepts. The results evaluated through both algorithmic metrics and human studies demonstrate that our CCRT can effectively remove the targeted concepts in a continuous manner while maintaining the high generation quality (e.g., text-image alignment) of the model.
Authors: Rui Liu, Shuwei He, Yifan Hu, Haizhou Li
Abstract: Visual Text-to-Speech (VTTS) aims to take the environmental image as the prompt to synthesize the reverberant speech for the spoken content. The challenge of this task lies in understanding the spatial environment from the image. Many attempts have been made to extract global spatial visual information from the RGB space of an spatial image. However, local and depth image information are crucial for understanding the spatial environment, which previous works have ignored. To address the issues, we propose a novel multi-modal and multi-scale spatial environment understanding scheme to achieve immersive VTTS, termed M2SE-VTTS. The multi-modal aims to take both the RGB and Depth spaces of the spatial image to learn more comprehensive spatial information, and the multi-scale seeks to model the local and global spatial knowledge simultaneously. Specifically, we first split the RGB and Depth images into patches and adopt the Gemini-generated environment captions to guide the local spatial understanding. After that, the multi-modal and multi-scale features are integrated by the local-aware global spatial understanding. In this way, M2SE-VTTS effectively models the interactions between local and global spatial contexts in the multi-modal spatial environment. Objective and subjective evaluations suggest that our model outperforms the advanced baselines in environmental speech generation. The code and audio samples are available at: https://github.com/AI-S2-Lab/M2SE-VTTS.
Authors: Chenfan Qu, Jian Liu, Haoxing Chen, Baihan Yu, Jingjing Liu, Weiqiang Wang, Lianwen Jin
Abstract: Recently, tampered text detection has attracted increasing attention due to its essential role in information security. Although existing methods can detect the tampered text region, the interpretation of such detection remains unclear, making the prediction unreliable. To address this problem, we propose to explain the basis of tampered text detection with natural language via large multimodal models. To fill the data gap for this task, we propose a large-scale, comprehensive dataset, ETTD, which contains both pixel-level annotations for tampered text region and natural language annotations describing the anomaly of the tampered text. Multiple methods are employed to improve the quality of the proposed data. For example, elaborate queries are introduced to generate high-quality anomaly descriptions with GPT4o. A fused mask prompt is proposed to reduce confusion when querying GPT4o to generate anomaly descriptions. To automatically filter out low-quality annotations, we also propose to prompt GPT4o to recognize tampered texts before describing the anomaly, and to filter out the responses with low OCR accuracy. To further improve explainable tampered text detection, we propose a simple yet effective model called TextSleuth, which achieves improved fine-grained perception and cross-domain generalization by focusing on the suspected region, with a two-stage analysis paradigm and an auxiliary grounding prompt. Extensive experiments on both the ETTD dataset and the public dataset have verified the effectiveness of the proposed methods. In-depth analysis is also provided to inspire further research. Our dataset and code will be open-source.
Authors: Xiangyue Zhang, Jianfang Li, Jiaxu Zhang, Ziqiang Dang, Jianqiang Ren, Liefeng Bo, Zhigang Tu
Abstract: A good co-speech motion generation cannot be achieved without a careful integration of common rhythmic motion and rare yet essential semantic motion. In this work, we propose SemTalk for holistic co-speech motion generation with frame-level semantic emphasis. Our key insight is to separately learn general motions and sparse motions, and then adaptively fuse them. In particular, rhythmic consistency learning is explored to establish rhythm-related base motion, ensuring a coherent foundation that synchronizes gestures with the speech rhythm. Subsequently, textit{semantic emphasis learning is designed to generate semantic-aware sparse motion, focusing on frame-level semantic cues. Finally, to integrate sparse motion into the base motion and generate semantic-emphasized co-speech gestures, we further leverage a learned semantic score for adaptive synthesis. Qualitative and quantitative comparisons on two public datasets demonstrate that our method outperforms the state-of-the-art, delivering high-quality co-speech motion with enhanced semantic richness over a stable base motion.
Authors: Chenxi Zhang, Qing Zhang, Jiayun Wu, Youwei Pang
Abstract: Camouflaged Object Detection (COD) aims to identify objects that blend seamlessly into their surroundings. The inherent visual complexity of camouflaged objects, including their low contrast with the background, diverse textures, and subtle appearance variations, often obscures semantic cues, making accurate segmentation highly challenging. Existing methods primarily rely on visual features, which are insufficient to handle the variability and intricacy of camouflaged objects, leading to unstable object perception and ambiguous segmentation results. To tackle these limitations, we introduce a novel task, class-guided camouflaged object detection (CGCOD), which extends traditional COD task by incorporating object-specific class knowledge to enhance detection robustness and accuracy. To facilitate this task, we present a new dataset, CamoClass, comprising real-world camouflaged objects with class annotations. Furthermore, we propose a multi-stage framework, CGNet, which incorporates a plug-and-play class prompt generator and a simple yet effective class-guided detector. This establishes a new paradigm for COD, bridging the gap between contextual understanding and class-guided detection. Extensive experimental results demonstrate the effectiveness of our flexible framework in improving the performance of proposed and existing detectors by leveraging class-level textual information.
Authors: Zechao Zhan, Dehong Gao, Jinxia Zhang, Jiale Huang, Yang Hu, Xin Wang
Abstract: Text-guided image editing model has achieved great success in general domain. However, directly applying these models to the fashion domain may encounter two issues: (1) Inaccurate localization of editing region; (2) Weak editing magnitude. To address these issues, the MADiff model is proposed. Specifically, to more accurately identify editing region, the MaskNet is proposed, in which the foreground region, densepose and mask prompts from large language model are fed into a lightweight UNet to predict the mask for editing region. To strengthen the editing magnitude, the Attention-Enhanced Diffusion Model is proposed, where the noise map, attention map, and the mask from MaskNet are fed into the proposed Attention Processor to produce a refined noise map. By integrating the refined noise map into the diffusion model, the edited image can better align with the target prompt. Given the absence of benchmarks in fashion image editing, we constructed a dataset named Fashion-E, comprising 28390 image-text pairs in the training set, and 2639 image-text pairs for four types of fashion tasks in the evaluation set. Extensive experiments on Fashion-E demonstrate that our proposed method can accurately predict the mask of editing region and significantly enhance editing magnitude in fashion image editing compared to the state-of-the-art methods.
Authors: Lihao Wang
Abstract: Surround-View System (SVS) is an essential component in Advanced Driver Assistance System (ADAS) and requires precise calibrations. However, conventional offline extrinsic calibration methods are cumbersome and time-consuming as they rely heavily on physical patterns. Additionally, these methods primarily focus on short-range areas surrounding the vehicle, resulting in lower calibration quality in more distant zones. To address these limitations, we propose Click-Calib, a pattern-free approach for offline SVS extrinsic calibration. Without requiring any special setup, the user only needs to click a few keypoints on the ground in natural scenes. Unlike other offline calibration approaches, Click-Calib optimizes camera poses over a wide range by minimizing reprojection distance errors of keypoints, thereby achieving accurate calibrations at both short and long distances. Furthermore, Click-Calib supports both single-frame and multiple-frame modes, with the latter offering even better results. Evaluations on our in-house dataset and the public WoodScape dataset demonstrate its superior accuracy and robustness compared to baseline methods. Code is available at https://github.com/lwangvaleo/click_calib.
Authors: Chaojie Mao, Jingfeng Zhang, Yulin Pan, Zeyinzi Jiang, Zhen Han, Yu Liu, Jingren Zhou
Abstract: We report ACE++, an instruction-based diffusion framework that tackles various image generation and editing tasks. Inspired by the input format for the inpainting task proposed by FLUX.1-Fill-dev, we improve the Long-context Condition Unit (LCU) introduced in ACE and extend this input paradigm to any editing and generation tasks. To take full advantage of image generative priors, we develop a two-stage training scheme to minimize the efforts of finetuning powerful text-to-image diffusion models like FLUX.1-dev. In the first stage, we pre-train the model using task data with the 0-ref tasks from the text-to-image model. There are many models in the community based on the post-training of text-to-image foundational models that meet this training paradigm of the first stage. For example, FLUX.1-Fill-dev deals primarily with painting tasks and can be used as an initialization to accelerate the training process. In the second stage, we finetune the above model to support the general instructions using all tasks defined in ACE. To promote the widespread application of ACE++ in different scenarios, we provide a comprehensive set of models that cover both full finetuning and lightweight finetuning, while considering general applicability and applicability in vertical scenarios. The qualitative analysis showcases the superiority of ACE++ in terms of generating image quality and prompt following ability. Code and models will be available on the project page: https://ali-vilab. github.io/ACE_plus_page/.
URLs: https://ali-vilab.
Authors: Chan Lee, Seungho Shin, Gyeong-Moon Park, Jung Uk Kim
Abstract: Although existing Sparsely Annotated Object Detection (SAOD) approches have made progress in handling sparsely annotated environments in multispectral domain, where only some pedestrians are annotated, they still have the following limitations: (i) they lack considerations for improving the quality of pseudo-labels for missing annotations, and (ii) they rely on fixed ground truth annotations, which leads to learning only a limited range of pedestrian visual appearances in the multispectral domain. To address these issues, we propose a novel framework called Sparsely Annotated Multispectral Pedestrian Detection (SAMPD). For limitation (i), we introduce Multispectral Pedestrian-aware Adaptive Weight (MPAW) and Positive Pseudo-label Enhancement (PPE) module. Utilizing multispectral knowledge, these modules ensure the generation of high-quality pseudo-labels and enable effective learning by increasing weights for high-quality pseudo-labels based on modality characteristics. To address limitation (ii), we propose an Adaptive Pedestrian Retrieval Augmentation (APRA) module, which adaptively incorporates pedestrian patches from ground-truth and dynamically integrates high-quality pseudo-labels with the ground-truth, facilitating a more diverse learning pool of pedestrians. Extensive experimental results demonstrate that our SAMPD significantly enhances performance in sparsely annotated environments within the multispectral domain.
Authors: Sun-Hyuk Choi, Hayoung Jo, Seong-Whan Lee
Abstract: Referring video object segmentation aims to segment objects within a video corresponding to a given text description. Existing transformer-based temporal modeling approaches face challenges related to query inconsistency and the limited consideration of context. Query inconsistency produces unstable masks of different objects in the middle of the video. The limited consideration of context leads to the segmentation of incorrect objects by failing to adequately account for the relationship between the given text and instances. To address these issues, we propose the Multi-context Temporal Consistency Module (MTCM), which consists of an Aligner and a Multi-Context Enhancer (MCE). The Aligner removes noise from queries and aligns them to achieve query consistency. The MCE predicts text-relevant queries by considering multi-context. We applied MTCM to four different models, increasing performance across all of them, particularly achieving 47.6 J&F on the MeViS. Code is available at https://github.com/Choi58/MTCM.
Authors: Xuyang Liu, Ziming Wang, Yuhang Han, Yingyao Wang, Jiale Yuan, Jun Song, Bo Zheng, Linfeng Zhang, Siteng Huang, Honggang Chen
Abstract: Multimodal large language models (MLLMs) have attracted considerable attention due to their exceptional performance in visual content understanding and reasoning. However, their inference efficiency has been a notable concern, as the increasing length of multimodal contexts leads to quadratic complexity. Token compression techniques, which reduce the number of visual tokens, have demonstrated their effectiveness in reducing computational costs. Yet, these approaches have struggled to keep pace with the rapid advancements in MLLMs, especially the AnyRes strategy in the context of high-resolution image understanding. In this paper, we propose a novel token compression method, GlobalCom$^2$, tailored for high-resolution MLLMs that receive both the thumbnail and multiple crops. GlobalCom$^2$ treats the tokens derived from the thumbnail as the "commander" of the entire token compression process, directing the allocation of retention ratios and the specific compression for each crop. In this way, redundant tokens are eliminated while important local details are adaptively preserved to the highest extent feasible. Empirical results across 10 benchmarks reveal that GlobalCom$^2$ achieves an optimal balance between performance and efficiency, and consistently outperforms state-of-the-art token compression methods with LLaVA-NeXT-7B/13B models. Our code is released at https://github.com/xuyang-liu16/GlobalCom2.
Authors: Aafaf Ridouan, Amine Bohi, Youssef Mourchid
Abstract: Pain management and severity detection are crucial for effective treatment, yet traditional self-reporting methods are subjective and may be unsuitable for non-verbal individuals (people with limited speaking skills). To address this limitation, we explore automated pain detection using facial expressions. Our study leverages deep learning techniques to improve pain assessment by analyzing facial images from the Pain Emotion Faces Database (PEMF). We propose two novel approaches1: (1) a hybrid ConvNeXt model combined with Long Short-Term Memory (LSTM) blocks to analyze video frames and predict pain presence, and (2) a Spatio-Temporal Graph Convolution Network (STGCN) integrated with LSTM to process landmarks from facial images for pain detection. Our work represents the first use of the PEMF dataset for binary pain classification and demonstrates the effectiveness of these models through extensive experimentation. The results highlight the potential of combining spatial and temporal features for enhanced pain detection, offering a promising advancement in objective pain assessment methodologies.
Authors: Yue Hu, Rong Liu, Meida Chen, Peter Beerel, Andrew Feng
Abstract: Achieving high-fidelity 3D reconstruction from monocular video remains challenging due to the inherent limitations of traditional methods like Structure-from-Motion (SfM) and monocular SLAM in accurately capturing scene details. While differentiable rendering techniques such as Neural Radiance Fields (NeRF) address some of these challenges, their high computational costs make them unsuitable for real-time applications. Additionally, existing 3D Gaussian Splatting (3DGS) methods often focus on photometric consistency, neglecting geometric accuracy and failing to exploit SLAM's dynamic depth and pose updates for scene refinement. We propose a framework integrating dense SLAM with 3DGS for real-time, high-fidelity dense reconstruction. Our approach introduces SLAM-Informed Adaptive Densification, which dynamically updates and densifies the Gaussian model by leveraging dense point clouds from SLAM. Additionally, we incorporate Geometry-Guided Optimization, which combines edge-aware geometric constraints and photometric consistency to jointly optimize the appearance and geometry of the 3DGS scene representation, enabling detailed and accurate SLAM mapping reconstruction. Experiments on the Replica and TUM-RGBD datasets demonstrate the effectiveness of our approach, achieving state-of-the-art results among monocular systems. Specifically, our method achieves a PSNR of 36.864, SSIM of 0.985, and LPIPS of 0.040 on Replica, representing improvements of 10.7%, 6.4%, and 49.4%, respectively, over the previous SOTA. On TUM-RGBD, our method outperforms the closest baseline by 10.2%, 6.6%, and 34.7% in the same metrics. These results highlight the potential of our framework in bridging the gap between photometric and geometric dense 3D scene representations, paving the way for practical and efficient monocular dense reconstruction.
Authors: Weizhi Li
Abstract: Unlike image classification and annotation, for which deep network models have achieved dominating superior performances compared to traditional computer vision algorithms, deep learning for automatic image segmentation still faces critical challenges. One of such hurdles is to obtain ground-truth segmentations as the training labels for deep network training. Especially when we study biomedical images, such as histopathological images (histo-images), it is unrealistic to ask for manual segmentation labels as the ground truth for training due to the fine image resolution as well as the large image size and complexity. In this paper, instead of relying on clean segmentation labels, we study whether and how integrating imperfect or noisy segmentation results from off-the-shelf segmentation algorithms may help achieve better segmentation results through a new Adaptive Noise-Tolerant Network (ANTN) model. We extend the noisy label deep learning to image segmentation with two novel aspects: (1) multiple noisy labels can be integrated into one deep learning model; (2) noisy segmentation modeling, including probabilistic parameters, is adaptive, depending on the given testing image appearance. Implementation of the new ANTN model on both the synthetic data and real-world histo-images demonstrates its effectiveness and superiority over off-the-shelf and other existing deep-learning-based image segmentation algorithms.
Authors: Lin Liu, Yutong Wang, Jiahao Chen, Jianfang Li, Tangli Xue, Longlong Li, Jianqiang Ren, Liefeng Bo
Abstract: This report introduces Make-A-Character 2, an advanced system for generating high-quality 3D characters from single portrait photographs, ideal for game development and digital human applications. Make-A-Character 2 builds upon its predecessor by incorporating several significant improvements for image-based head generation. We utilize the IC-Light method to correct non-ideal illumination in input photos and apply neural network-based color correction to harmonize skin tones between the photos and game engine renders. We also employ the Hierarchical Representation Network to capture high-frequency facial structures and conduct adaptive skeleton calibration for accurate and expressive facial animations. The entire image-to-3D-character generation process takes less than 2 minutes. Furthermore, we leverage transformer architecture to generate co-speech facial and gesture actions, enabling real-time conversation with the generated character. These technologies have been integrated into our conversational AI avatar products.
Authors: Roi Papo, Sapir Gershov, Tom Friedman, Itay Or, Gil Bolotin, Shlomi Laufer
Abstract: Hand-specific localization has garnered significant interest within the computer vision community. Although there are numerous datasets with hand annotations from various angles and settings, domain transfer techniques frequently struggle in surgical environments. This is mainly due to the limited availability of gloved hand instances and the unique challenges of operating rooms (ORs). Thus, hand-detection models tailored to OR settings require extensive training and expensive annotation processes. To overcome these challenges, we present "RoHan" - a novel approach for robust hand detection in the OR, leveraging advanced semi-supervised domain adaptation techniques to tackle the challenges of varying recording conditions, diverse glove colors, and occlusions common in surgical settings. Our methodology encompasses two main stages: (1) data augmentation strategy that utilizes "Artificial Gloves," a method for augmenting publicly available hand datasets with synthetic images of hands-wearing gloves; (2) semi-supervised domain adaptation pipeline that improves detection performance in real-world OR settings through iterative prediction refinement and efficient frame filtering. We evaluate our method using two datasets: simulated enterotomy repair and saphenous vein graft harvesting. "RoHan" substantially reduces the need for extensive labeling and model training, paving the way for the practical implementation of hand detection technologies in medical settings.
Authors: Maura Pintor, Daniele Angioni, Angelo Sotgiu, Luca Demetrio, Ambra Demontis, Battista Biggio, Fabio Roli
Abstract: Adversarial patches are optimized contiguous pixel blocks in an input image that cause a machine-learning model to misclassify it. However, their optimization is computationally demanding, and requires careful hyperparameter tuning, potentially leading to suboptimal robustness evaluations. To overcome these issues, we propose ImageNet-Patch, a dataset to benchmark machine-learning models against adversarial patches. It consists of a set of patches, optimized to generalize across different models, and readily applicable to ImageNet data after preprocessing them with affine transformations. This process enables an approximate yet faster robustness evaluation, leveraging the transferability of adversarial perturbations. We showcase the usefulness of this dataset by testing the effectiveness of the computed patches against 127 models. We conclude by discussing how our dataset could be used as a benchmark for robustness, and how our methodology can be generalized to other domains. We open source our dataset and evaluation code at https://github.com/pralab/ImageNet-Patch.
Authors: Xinyuan Song
Abstract: With the spread of COVID-19 around the globe over the past year, the usage of artificial intelligence (AI) algorithms and image processing methods to analyze the X-ray images of patients' chest with COVID-19 has become essential. The COVID-19 virus recognition in the lung area of a patient is one of the basic and essential needs of clicical centers and hospitals. Most research in this field has been devoted to papers on the basis of deep learning methods utilizing CNNs (Convolutional Neural Network), which mainly deal with the screening of sick and healthy people.In this study, a new structure of a 19-layer CNN has been recommended for accurately recognition of the COVID-19 from the X-ray pictures of chest. The offered CNN is developed to serve as a precise diagnosis system for a three class (viral pneumonia, Normal, COVID) and a four classclassification (Lung opacity, Normal, COVID-19, and pneumonia). A comparison is conducted among the outcomes of the offered procedure and some popular pretrained networks, including Inception, Alexnet, ResNet50, Squeezenet, and VGG19 and based on Specificity, Accuracy, Precision, Sensitivity, Confusion Matrix, and F1-score. The experimental results of the offered CNN method specify its dominance over the existing published procedures. This method can be a useful tool for clinicians in deciding properly about COVID-19.
Authors: Sadanand Modak, Noah Patton, Isil Dillig, Joydeep Biswas
Abstract: This paper addresses the problem of preference learning, which aims to align robot behaviors through learning user specific preferences (e.g. "good pull-over location") from visual demonstrations. Despite its similarity to learning factual concepts (e.g. "red door"), preference learning is a fundamentally harder problem due to its subjective nature and the paucity of person-specific training data. We address this problem using a novel framework called SYNAPSE, which is a neuro-symbolic approach designed to efficiently learn preferential concepts from limited data. SYNAPSE represents preferences as neuro-symbolic programs, facilitating inspection of individual parts for alignment, in a domain-specific language (DSL) that operates over images and leverages a novel combination of visual parsing, large language models, and program synthesis to learn programs representing individual preferences. We perform extensive evaluations on various preferential concepts as well as user case studies demonstrating its ability to align well with dissimilar user preferences. Our method significantly outperforms baselines, especially when it comes to out of distribution generalization. We show the importance of the design choices in the framework through multiple ablation studies. Code, additional results, and supplementary material can be found on the website: https://amrl.cs.utexas.edu/synapse
Authors: Tony Lindeberg
Abstract: This paper presents an analysis of properties of two hybrid discretization methods for Gaussian derivatives, based on convolutions with either the normalized sampled Gaussian kernel or the integrated Gaussian kernel followed by central differences. The motivation for studying these discretization methods is that in situations when multiple spatial derivatives of different order are needed at the same scale level, they can be computed significantly more efficiently compared to more direct derivative approximations based on explicit convolutions with either sampled Gaussian kernels or integrated Gaussian kernels. While these computational benefits do also hold for the genuinely discrete approach for computing discrete analogues of Gaussian derivatives, based on convolution with the discrete analogue of the Gaussian kernel followed by central differences, the underlying mathematical primitives for the discrete analogue of the Gaussian kernel, in terms of modified Bessel functions of integer order, may not be available in certain frameworks for image processing, such as when performing deep learning based on scale-parameterized filters in terms of Gaussian derivatives, with learning of the scale levels. In this paper, we present a characterization of the properties of these hybrid discretization methods, in terms of quantitative performance measures concerning the amount of spatial smoothing that they imply, as well as the relative consistency of scale estimates obtained from scale-invariant feature detectors with automatic scale selection, with an emphasis on the behaviour for very small values of the scale parameter, which may differ significantly from corresponding results obtained from the fully continuous scale-space theory, as well as between different types of discretization methods.
Authors: Randall Balestriero, Ahmed Imtiaz Humayun, Richard Baraniuk
Abstract: In this paper, we overview one promising avenue of progress at the mathematical foundation of deep learning: the connection between deep networks and function approximation by affine splines (continuous piecewise linear functions in multiple dimensions). In particular, we will overview work over the past decade on understanding certain geometrical properties of a deep network's affine spline mapping, in particular how it tessellates its input space. As we will see, the affine spline connection and geometrical viewpoint provide a powerful portal through which to view, analyze, and improve the inner workings of a deep network.
Authors: Yao Ni, Shan Zhang, Piotr Koniusz
Abstract: Parameter-Efficient Fine-Tuning (PEFT) effectively adapts pre-trained transformers to downstream tasks. However, the optimization of tasks performance often comes at the cost of generalizability in fine-tuned models. To address this issue, we theoretically connect smaller weight gradient norms during training and larger datasets to the improvements in model generalization. Motivated by this connection, we propose reducing gradient norms for enhanced generalization and aligning fine-tuned model with the pre-trained counterpart to retain knowledge from large-scale pre-training data. Yet, naive alignment does not guarantee gradient reduction and can potentially cause gradient explosion, complicating efforts to manage gradients. To address such an issue, we propose PACE, marrying generalization of PArameter-efficient fine-tuning with Consistency rEgularization. We perturb features learned from the adapter with the multiplicative noise and ensure the fine-tuned model remains consistent for same sample under different perturbations. Theoretical analysis shows that PACE not only implicitly regularizes gradients for enhanced generalization, but also implicitly aligns the fine-tuned and pre-trained models to retain knowledge. Experimental evidence supports our theories. PACE surpasses existing PEFT methods in visual adaptation tasks (VTAB-1k, FGVC, few-shot learning, domain adaptation) showcasing its potential for resource-efficient fine-tuning. It also improves LoRA in text classification (GLUE) and mathematical reasoning (GSM-8K). The code is available at https://github.com/MaxwellYaoNi/PACE
Authors: Jonathan von Rad, Florian Seuffert
Abstract: Deep Neural Networks (DNNs) are often over-parameterized for their tasks and can be compressed quite drastically by removing weights, a process called pruning. We investigate the impact of different pruning techniques on the classification performance and interpretability of GoogLeNet. We systematically apply unstructured and structured pruning, as well as connection sparsity (pruning of input weights) methods to the network and analyze the outcomes regarding the network's performance on the validation set of ImageNet. We also compare different retraining strategies, such as iterative pruning and one-shot pruning. We find that with sufficient retraining epochs, the performance of the networks can approximate the performance of the default GoogLeNet - and even surpass it in some cases. To assess interpretability, we employ the Mechanistic Interpretability Score (MIS) developed by Zimmermann et al. . Our experiments reveal that there is no significant relationship between interpretability and pruning rate when using MIS as a measure. Additionally, we observe that networks with extremely low accuracy can still achieve high MIS scores, suggesting that the MIS may not always align with intuitive notions of interpretability, such as understanding the basis of correct decisions.
Authors: Jean-Eudes Ayilo (MULTISPEECH), Mostafa Sadeghi (MULTISPEECH), Romain Serizel (MULTISPEECH), Xavier Alameda-Pineda (ROBOTLEARN)
Abstract: This paper proposes a new unsupervised audio-visual speech enhancement (AVSE) approach that combines a diffusion-based audio-visual speech generative model with a non-negative matrix factorization (NMF) noise model. First, the diffusion model is pre-trained on clean speech conditioned on corresponding video data to simulate the speech generative distribution. This pre-trained model is then paired with the NMF-based noise model to estimate clean speech iteratively. Specifically, a diffusion-based posterior sampling approach is implemented within the reverse diffusion process, where after each iteration, a speech estimate is obtained and used to update the noise parameters. Experimental results confirm that the proposed AVSE approach not only outperforms its audio-only counterpart but also generalizes better than a recent supervised-generative AVSE method. Additionally, the new inference algorithm offers a better balance between inference speed and performance compared to the previous diffusion-based method. Code and demo available at: https://jeaneudesayilo.github.io/fast_UdiffSE
Authors: John Brandon Graham-Knight, Jamil Fayyad, Nourhan Bayasi, Patricia Lasserre, Homayoun Najjaran
Abstract: Class imbalance and label noise are pervasive in large-scale datasets, yet much of machine learning research assumes well-labeled, balanced data, which rarely reflects real world conditions. Existing approaches typically address either label noise or class imbalance in isolation, leading to suboptimal results when both issues coexist. In this work, we propose Conformal-in-the-Loop (CitL), a novel training framework that addresses both challenges with a conformal prediction-based approach. CitL evaluates sample uncertainty to adjust weights and prune unreliable examples, enhancing model resilience and accuracy with minimal computational cost. Our extensive experiments include a detailed analysis showing how CitL effectively emphasizes impactful data in noisy, imbalanced datasets. Our results show that CitL consistently boosts model performance, achieving up to a 6.1% increase in classification accuracy and a 5.0 mIoU improvement in segmentation. Our code is publicly available: CitL.
Authors: Moritz Schneider, Robert Krug, Narunas Vaskevicius, Luigi Palmieri, Joschka Boedecker
Abstract: Visual Reinforcement Learning (RL) methods often require extensive amounts of data. As opposed to model-free RL, model-based RL (MBRL) offers a potential solution with efficient data utilization through planning. Additionally, RL lacks generalization capabilities for real-world tasks. Prior work has shown that incorporating pre-trained visual representations (PVRs) enhances sample efficiency and generalization. While PVRs have been extensively studied in the context of model-free RL, their potential in MBRL remains largely unexplored. In this paper, we benchmark a set of PVRs on challenging control tasks in a model-based RL setting. We investigate the data efficiency, generalization capabilities, and the impact of different properties of PVRs on the performance of model-based agents. Our results, perhaps surprisingly, reveal that for MBRL current PVRs are not more sample efficient than learning representations from scratch, and that they do not generalize better to out-of-distribution (OOD) settings. To explain this, we analyze the quality of the trained dynamics model. Furthermore, we show that data diversity and network architecture are the most important contributors to OOD generalization performance.
Authors: Alexis Fox, Samarth Swarup, Abhijin Adiga
Abstract: Considering the difficulty of interpreting generative model output, there is significant current research focused on determining meaningful evaluation metrics. Several recent approaches utilize "precision" and "recall," borrowed from the classification domain, to individually quantify the output fidelity (realism) and output diversity (representation of the real data variation), respectively. With the increase in metric proposals, there is a need for a unifying perspective, allowing for easier comparison and clearer explanation of their benefits and drawbacks. To this end, we unify a class of kth-nearest-neighbors (kNN)-based metrics under an information-theoretic lens using approaches from kNN density estimation. Additionally, we propose a tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity and two distinct aspects of diversity, inter- and intra-class. Our domain-agnostic metric, derived from the information-theoretic concepts of entropy and cross-entropy, can be dissected for both sample- and mode-level analysis. Our detailed experimental results demonstrate the sensitivity of our metric components to their respective qualities and reveal undesirable behaviors of other metrics.
Authors: Claire Huchthausen, Menglin Shi, Gabriel L. A. de Sousa, Jonathan Colen, Emery Shelley, James Larner, Einsley Janowski, Krishni Wijesooriya
Abstract: BACKGROUND: Radiomics provides quantitative features of pulmonary nodules (PNs) which could aid lung cancer diagnosis, but medical image acquisition variability is an obstacle to clinical application. Acquisition effects may differ between radiomic features from benign vs. malignant PNs. PURPOSE: We evaluated how to account for differences between benign and malignant PNs when correcting radiomic features' acquisition dependency. METHODS: We used 567 chest CT scans grouped as benign, malignant, or lung cancer screening (mixed benign, malignant). ComBat harmonization was applied to extracted features for variation in 4 acquisition parameters. We compared: harmonizing without distinction, harmonizing with a covariate to preserve distinctions between subgroups, and harmonizing subgroups separately. Significant ($p\le0.05$) Kruskal-Wallis tests showed whether harmonization removed acquisition dependency. A LASSO-SVM pipeline was trained on successfully harmonized features to predict malignancy. To evaluate predictive information in these features, the trained harmonization estimators and predictive model were applied to unseen test sets. Harmonization and predictive performance were assessed for 10 trials of 5-fold cross-validation. RESULTS: An average 2.1% of features (95% CI:1.9-2.4%) were acquisition-independent when harmonized without distinction, 27.3% (95% CI:25.7-28.9%) when harmonized with a covariate, and 90.9% (95% CI:90.4-91.5%) when harmonized separately. Data harmonized separately or with a covariate trained models with higher ROC-AUC for screening scans than data harmonized without distinction between benign and malignant PNs (Delong test, adjusted $p\le0.05$). CONCLUSIONS: Radiomic features of benign and malignant PNs need different corrective transformations to recover acquisition-independent distributions. This can be done by harmonizing separately or with a covariate.
Authors: Chenyu You, Haocheng Dai, Yifei Min, Jasjeet S. Sekhon, Sarang Joshi, James S. Duncan
Abstract: Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on \textit{atypical} examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we articulate the hypothesis: the imbalanced group performance is a byproduct of ``noisy'' spurious memorization confined to a small set of neurons. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation.
Authors: Raghav Singhal, Zachary Horvitz, Ryan Teehan, Mengye Ren, Zhou Yu, Kathleen McKeown, Rajesh Ranganath
Abstract: Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
URLs: https://github.com/zacharyhorvitz/Fk-Diffusion-Steering
Authors: Marios Aristodemou, Xiaolan Liu, Yuan Wang, Konstantinos G. Kyriakopoulos, Sangarapillai Lambotharan, Qingsong Wei
Abstract: As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.