Authors: Owen Oertell, Jonathan D. Chang, Yiyi Zhang, Kiant\'e Brantley, Wen Sun
Abstract: Reinforcement learning (RL) has improved guided image generation with diffusion models by directly optimizing rewards that capture image quality, aesthetics, and instruction following capabilities. However, the resulting generative policies inherit the same iterative sampling process of diffusion models that causes slow generation. To overcome this limitation, consistency models proposed learning a new class of generative models that directly map noise to data, resulting in a model that can generate an image in as few as one sampling iteration. In this work, to optimize text-to-image generative models for task specific rewards and enable fast training and inference, we propose a framework for fine-tuning consistency models via RL. Our framework, called Reinforcement Learning for Consistency Model (RLCM), frames the iterative inference process of a consistency model as an RL procedure. RLCM improves upon RL fine-tuned diffusion models on text-to-image generation capabilities and trades computation during inference time for sample quality. Experimentally, we show that RLCM can adapt text-to-image consistency models to objectives that are challenging to express with prompting, such as image compressibility, and those derived from human feedback, such as aesthetic quality. Comparing to RL finetuned diffusion models, RLCM trains significantly faster, improves the quality of the generation measured under the reward objectives, and speeds up the inference procedure by generating high quality images with as few as two inference steps. Our code is available at https://rlcm.owenoertell.com
Authors: Zijie Wu, Chaohui Yu, Yanqin Jiang, Chenjie Cao, Fan Wang, Xiang Bai
Abstract: Recent advances in 2D/3D generative models enable the generation of dynamic 3D objects from a single-view video. Existing approaches utilize score distillation sampling to form the dynamic scene as dynamic NeRF or dense 3D Gaussians. However, these methods struggle to strike a balance among reference view alignment, spatio-temporal consistency, and motion fidelity under single-view conditions due to the implicit nature of NeRF or the intricate dense Gaussian motion prediction. To address these issues, this paper proposes an efficient, sparse-controlled video-to-4D framework named SC4D, that decouples motion and appearance to achieve superior video-to-4D generation. Moreover, we introduce Adaptive Gaussian (AG) initialization and Gaussian Alignment (GA) loss to mitigate shape degeneration issue, ensuring the fidelity of the learned motion and shape. Comprehensive experimental results demonstrate that our method surpasses existing methods in both quality and efficiency. In addition, facilitated by the disentangled modeling of motion and appearance of SC4D, we devise a novel application that seamlessly transfers the learned motion onto a diverse array of 4D entities according to textual descriptions.
Authors: Alex Costanzino, Pierluigi Zama Ramirez, Mirko Del Moro, Agostino Aiezzo, Giuseppe Lisanti, Samuele Salti, Luigi Di Stefano
Abstract: Anomaly Detection and Segmentation (AD&S) is crucial for industrial quality control. While existing methods excel in generating anomaly scores for each pixel, practical applications require producing a binary segmentation to identify anomalies. Due to the absence of labeled anomalies in many real scenarios, standard practices binarize these maps based on some statistics derived from a validation set containing only nominal samples, resulting in poor segmentation performance. This paper addresses this problem by proposing a test time training strategy to improve the segmentation performance. Indeed, at test time, we can extract rich features directly from anomalous samples to train a classifier that can discriminate defects effectively. Our general approach can work downstream to any AD&S method that provides an anomaly score map as output, even in multimodal settings. We demonstrate the effectiveness of our approach over baselines through extensive experimentation and evaluation on MVTec AD and MVTec 3D-AD.
Authors: Simon Weber, Bar{\i}\c{s} Z\"ong\"ur, Nikita Araslanov, Daniel Cremers
Abstract: Hierarchy is a natural representation of semantic taxonomies, including the ones routinely used in image segmentation. Indeed, recent work on semantic segmentation reports improved accuracy from supervised training leveraging hierarchical label structures. Encouraged by these results, we revisit the fundamental assumptions behind that work. We postulate and then empirically verify that the reasons for the observed improvement in segmentation accuracy may be entirely unrelated to the use of the semantic hierarchy. To demonstrate this, we design a range of cross-domain experiments with a representative hierarchical approach. We find that on the new testing domains, a flat (non-hierarchical) segmentation network, in which the parents are inferred from the children, has superior segmentation accuracy to the hierarchical approach across the board. Complementing these findings and inspired by the intrinsic properties of hyperbolic spaces, we study a more principled approach to hierarchical segmentation using the Poincar\'e ball model. The hyperbolic representation largely outperforms the previous (Euclidean) hierarchical approach as well and is on par with our flat Euclidean baseline in terms of segmentation accuracy. However, it additionally exhibits surprisingly strong calibration quality of the parent nodes in the semantic hierarchy, especially on the more challenging domains. Our combined analysis suggests that the established practice of hierarchical segmentation may be limited to in-domain settings, whereas flat classifiers generalize substantially better, especially if they are modeled in the hyperbolic space.
Authors: Juanwu Lu, Can Cui, Yunsheng Ma, Aniket Bera, Ziran Wang
Abstract: Safety and robustness are crucial factors in developing trustworthy autonomous vehicles. One essential aspect of addressing these factors is to equip vehicles with the capability to predict future trajectories for all moving objects in the surroundings and quantify prediction uncertainties. In this paper, we propose the Sequential Neural Variational Agent (SeNeVA), a generative model that describes the distribution of future trajectories for a single moving object. Our approach can distinguish Out-of-Distribution data while quantifying uncertainty and achieving competitive performance compared to state-of-the-art methods on the Argoverse 2 and INTERACTION datasets. Specifically, a 0.446 meters minimum Final Displacement Error, a 0.203 meters minimum Average Displacement Error, and a 5.35% Miss Rate are achieved on the INTERACTION test set. Extensive qualitative and quantitative analysis is also provided to evaluate the proposed model. Our open-source code is available at https://github.com/PurdueDigitalTwin/seneva.
Authors: Elham Amin Mansour, Ozan Unal, Suman Saha, Benjamin Bejar, Luc Van Gool
Abstract: The increasing relevance of panoptic segmentation is tied to the advancements in autonomous driving and AR/VR applications. However, the deployment of such models has been limited due to the expensive nature of dense data annotation, giving rise to unsupervised domain adaptation (UDA). A key challenge in panoptic UDA is reducing the domain gap between a labeled source and an unlabeled target domain while harmonizing the subtasks of semantic and instance segmentation to limit catastrophic interference. While considerable progress has been achieved, existing approaches mainly focus on the adaptation of semantic segmentation. In this work, we focus on incorporating instance-level adaptation via a novel instance-aware cross-domain mixing strategy IMix. IMix significantly enhances the panoptic quality by improving instance segmentation performance. Specifically, we propose inserting high-confidence predicted instances from the target domain onto source images, retaining the exhaustiveness of the resulting pseudo-labels while reducing the injected confirmation bias. Nevertheless, such an enhancement comes at the cost of degraded semantic performance, attributed to catastrophic forgetting. To mitigate this issue, we regularize our semantic branch by employing CLIP-based domain alignment (CDA), exploiting the domain-robustness of natural language prompts. Finally, we present an end-to-end model incorporating these two mechanisms called LIDAPS, achieving state-of-the-art results on all popular panoptic UDA benchmarks.
Authors: Qinji Yu, Yirui Wang, Ke Yan, Haoshen Li, Dazhou Guo, Li Zhang, Le Lu, Na Shen, Qifeng Wang, Xiaowei Ding, Xianghua Ye, Dakai Jin
Abstract: Lymph node (LN) assessment is a critical, indispensable yet very challenging task in the routine clinical workflow of radiology and oncology. Accurate LN analysis is essential for cancer diagnosis, staging, and treatment planning. Finding scatteredly distributed, low-contrast clinically relevant LNs in 3D CT is difficult even for experienced physicians under high inter-observer variations. Previous automatic LN detection works typically yield limited recall and high false positives (FPs) due to adjacent anatomies with similar image intensities, shapes, or textures (vessels, muscles, esophagus, etc). In this work, we propose a new LN DEtection TRansformer, named LN-DETR, to achieve more accurate performance. By enhancing the 2D backbone with a multi-scale 2.5D feature fusion to incorporate 3D context explicitly, more importantly, we make two main contributions to improve the representation quality of LN queries. 1) Considering that LN boundaries are often unclear, an IoU prediction head and a location debiased query selection are proposed to select LN queries of higher localization accuracy as the decoder query's initialization. 2) To reduce FPs, query contrastive learning is employed to explicitly reinforce LN queries towards their best-matched ground-truth queries over unmatched query predictions. Trained and tested on 3D CT scans of 1067 patients (with 10,000+ labeled LNs) via combining seven LN datasets from different body parts (neck, chest, and abdomen) and pathologies/cancers, our method significantly improves the performance of previous leading methods by > 4-5% average recall at the same FP rates in both internal and external testing. We further evaluate on the universal lesion detection task using NIH DeepLesion benchmark, and our method achieves the top performance of 88.46% averaged recall across 0.5 to 4 FPs per image, compared with other leading reported results.
Authors: Jonathan F. Carter, Jo\~ao Jorge, Oliver Gibson, Lionel Tarassenko
Abstract: Advances in camera-based physiological monitoring have enabled the robust, non-contact measurement of respiration and the cardiac pulse, which are known to be indicative of the sleep stage. This has led to research into camera-based sleep monitoring as a promising alternative to "gold-standard" polysomnography, which is cumbersome, expensive to administer, and hence unsuitable for longer-term clinical studies. In this paper, we introduce SleepVST, a transformer model which enables state-of-the-art performance in camera-based sleep stage classification (sleep staging). After pre-training on contact sensor data, SleepVST outperforms existing methods for cardio-respiratory sleep staging on the SHHS and MESA datasets, achieving total Cohen's kappa scores of 0.75 and 0.77 respectively. We then show that SleepVST can be successfully transferred to cardio-respiratory waveforms extracted from video, enabling fully contact-free sleep staging. Using a video dataset of 50 nights, we achieve a total accuracy of 78.8\% and a Cohen's $\kappa$ of 0.71 in four-class video-based sleep staging, setting a new state-of-the-art in the domain.
Authors: Amrin Kareem, Jean Lahoud, Hisham Cholakkal
Abstract: Recent advancements in 3D perception systems have significantly improved their ability to perform visual recognition tasks such as segmentation. However, these systems still heavily rely on explicit human instruction to identify target objects or categories, lacking the capability to actively reason and comprehend implicit user intentions. We introduce a novel segmentation task known as reasoning part segmentation for 3D objects, aiming to output a segmentation mask based on complex and implicit textual queries about specific parts of a 3D object. To facilitate evaluation and benchmarking, we present a large 3D dataset comprising over 60k instructions paired with corresponding ground-truth part segmentation annotations specifically curated for reasoning-based 3D part segmentation. We propose a model that is capable of segmenting parts of 3D objects based on implicit textual queries and generating natural language explanations corresponding to 3D object segmentation requests. Experiments show that our method achieves competitive performance to models that use explicit queries, with the additional abilities to identify part concepts, reason about them, and complement them with world knowledge. Our source code, dataset, and trained models are available at https://github.com/AmrinKareem/PARIS3D.
Authors: Gianluca Barone, Aashrit Cunchala, Rudy Nunez
Abstract: Standard classification theory assumes that the distribution of images in the test and training sets are identical. Unfortunately, real-life scenarios typically feature unseen data ("out-of-distribution data") which is different from data in the training distribution("in-distribution"). This issue is most prevalent in social justice problems where data from under-represented groups may appear in the test data without representing an equal proportion of the training data. This may result in a model returning confidently wrong decisions and predictions. We are interested in the following question: Can the performance of a neural network improve on facial images of out-of-distribution data when it is trained simultaneously on multiple datasets of in-distribution data? We approach this problem by incorporating the Outlier Exposure model and investigate how the model's performance changes when other datasets of facial images were implemented. We observe that the accuracy and other metrics of the model can be increased by applying Outlier Exposure, incorporating a trainable weight parameter to increase the machine's emphasis on outlier images, and by re-weighting the importance of different class labels. We also experimented with whether sorting the images and determining outliers via image features would have more of an effect on the metrics than sorting by average pixel value. Our goal was to make models not only more accurate but also more fair by scanning a more expanded range of images. We also tested the datasets in reverse order to see whether a more fair dataset with balanced features has an effect on the model's accuracy.
Authors: Maryam Ahmed, Tooba Bibi, Rizwan Ahmed Khan, Sidra Nasir
Abstract: The study introduces an integrated framework combining Convolutional Neural Networks (CNNs) and Explainable Artificial Intelligence (XAI) for the enhanced diagnosis of breast cancer using the CBIS-DDSM dataset. Utilizing a fine-tuned ResNet50 architecture, our investigation not only provides effective differentiation of mammographic images into benign and malignant categories but also addresses the opaque "black-box" nature of deep learning models by employing XAI methodologies, namely Grad-CAM, LIME, and SHAP, to interpret CNN decision-making processes for healthcare professionals. Our methodology encompasses an elaborate data preprocessing pipeline and advanced data augmentation techniques to counteract dataset limitations, and transfer learning using pre-trained networks, such as VGG-16, DenseNet and ResNet was employed. A focal point of our study is the evaluation of XAI's effectiveness in interpreting model predictions, highlighted by utilising the Hausdorff measure to assess the alignment between AI-generated explanations and expert annotations quantitatively. This approach plays a critical role for XAI in promoting trustworthiness and ethical fairness in AI-assisted diagnostics. The findings from our research illustrate the effective collaboration between CNNs and XAI in advancing diagnostic methods for breast cancer, thereby facilitating a more seamless integration of advanced AI technologies within clinical settings. By enhancing the interpretability of AI-driven decisions, this work lays the groundwork for improved collaboration between AI systems and medical practitioners, ultimately enriching patient care. Furthermore, the implications of our research extend well beyond the current methodologies, advocating for subsequent inquiries into the integration of multimodal data and the refinement of AI explanations to satisfy the needs of clinical practice.
Authors: Anas Mohammad Ishfaqul Muktadir Osmani, Taimur Rahman, Salekul Islam
Abstract: In this paper, we analyze the effectiveness of transfer learning on classifying electronic components. Transfer learning reuses pre-trained models to save time and resources in building a robust classifier rather than learning from scratch. Our work introduces a lightweight CNN, coined as VoltaVision, and compares its performance against more complex models. We test the hypothesis that transferring knowledge from a similar task to our target domain yields better results than state-of-the-art models trained on general datasets. Our dataset and code for this work are available at https://github.com/AnasIshfaque/VoltaVision.
Authors: Gihyun Kwon, Simon Jenni, Dingzeyu Li, Joon-Young Lee, Jong Chul Ye, Fabian Caba Heilbron
Abstract: While there has been significant progress in customizing text-to-image generation models, generating images that combine multiple personalized concepts remains challenging. In this work, we introduce Concept Weaver, a method for composing customized text-to-image diffusion models at inference time. Specifically, the method breaks the process into two steps: creating a template image aligned with the semantics of input prompts, and then personalizing the template using a concept fusion strategy. The fusion strategy incorporates the appearance of the target concepts into the template image while retaining its structural details. The results indicate that our method can generate multiple custom concepts with higher identity fidelity compared to alternative approaches. Furthermore, the method is shown to seamlessly handle more than two concepts and closely follow the semantic meaning of the input prompt without blending appearances across different subjects.
Authors: Manjin Kim, Paul Hongsuck Seo, Cordelia Schmid, Minsu Cho
Abstract: We introduce a new attention mechanism, dubbed structural self-attention (StructSA), that leverages rich correlation patterns naturally emerging in key-query interactions of attention. StructSA generates attention maps by recognizing space-time structures of key-query correlations via convolution and uses them to dynamically aggregate local contexts of value features. This effectively leverages rich structural patterns in images and videos such as scene layouts, object motion, and inter-object relations. Using StructSA as a main building block, we develop the structural vision transformer (StructViT) and evaluate its effectiveness on both image and video classification tasks, achieving state-of-the-art results on ImageNet-1K, Kinetics-400, Something-Something V1 & V2, Diving-48, and FineGym.
Authors: Xuecan Wang, Shibang Xiao, Xiaohui Liang
Abstract: We present a lightweight solution for estimating spatially-coherent indoor lighting from a single RGB image. Previous methods for estimating illumination using volumetric representations have overlooked the sparse distribution of light sources in space, necessitating substantial memory and computational resources for achieving high-quality results. We introduce a unified, voxel octree-based illumination estimation framework to produce 3D spatially-coherent lighting. Additionally, a differentiable voxel octree cone tracing rendering layer is proposed to eliminate regular volumetric representation throughout the entire process and ensure the retention of features across different frequency domains. This reduction significantly decreases spatial usage and required floating-point operations without substantially compromising precision. Experimental results demonstrate that our approach achieves high-quality coherent estimation with minimal cost compared to previous methods.
Authors: Lynn Miller, Charlotte Pelletier, Geoffrey I. Webb
Abstract: Earth observation (EO) satellite missions have been providing detailed images about the state of the Earth and its land cover for over 50 years. Long term missions, such as NASA's Landsat, Terra, and Aqua satellites, and more recently, the ESA's Sentinel missions, record images of the entire world every few days. Although single images provide point-in-time data, repeated images of the same area, or satellite image time series (SITS) provide information about the changing state of vegetation and land use. These SITS are useful for modeling dynamic processes and seasonal changes such as plant phenology. They have potential benefits for many aspects of land and natural resource management, including applications in agricultural, forest, water, and disaster management, urban planning, and mining. However, the resulting satellite image time series (SITS) are complex, incorporating information from the temporal, spatial, and spectral dimensions. Therefore, deep learning methods are often deployed as they can analyze these complex relationships. This review presents a summary of the state-of-the-art methods of modelling environmental, agricultural, and other Earth observation variables from SITS data using deep learning methods. We aim to provide a resource for remote sensing experts interested in using deep learning techniques to enhance Earth observation models with temporal information.
Authors: Xingyu Liu, Chenyangguang Zhang, Gu Wang, Ruida Zhang, Xiangyang Ji
Abstract: In robotic vision, a de-facto paradigm is to learn in simulated environments and then transfer to real-world applications, which poses an essential challenge in bridging the sim-to-real domain gap. While mainstream works tackle this problem in the RGB domain, we focus on depth data synthesis and develop a range-aware RGB-D data simulation pipeline (RaSim). In particular, high-fidelity depth data is generated by imitating the imaging principle of real-world sensors. A range-aware rendering strategy is further introduced to enrich data diversity. Extensive experiments show that models trained with RaSim can be directly applied to real-world scenarios without any finetuning and excel at downstream RGB-D perception tasks.
Authors: Reina Kaneko, Hiroshi Higashi, Yuichi Tanaka
Abstract: This paper introduces the physics-inspired synthesized underwater image dataset (PHISWID), a dataset tailored for enhancing underwater image processing through physics-inspired image synthesis. Deep learning approaches to underwater image enhancement typically demand extensive datasets, yet acquiring paired clean and degraded underwater ones poses significant challenges. While several underwater image datasets have been proposed using physics-based synthesis, a publicly accessible collection has been lacking. Additionally, most underwater image synthesis approaches do not intend to reproduce atmospheric scenes, resulting in incomplete enhancement. PHISWID addresses this gap by offering a set of paired ground-truth (atmospheric) and synthetically degraded underwater images, showcasing not only color degradation but also the often-neglected effects of marine snow, a composite of organic matter and sand particles that considerably impairs underwater image clarity. The dataset applies these degradations to atmospheric RGB-D images, enhancing the dataset's realism and applicability. PHISWID is particularly valuable for training deep neural networks in a supervised learning setting and for objectively assessing image quality in benchmark analyses. Our results reveal that even a basic U-Net architecture, when trained with PHISWID, substantially outperforms existing methods in underwater image enhancement. We intend to release PHISWID publicly, contributing a significant resource to the advancement of underwater imaging technology.
Authors: Simon Weber, Thomas Dag\`es, Maolin Gao, Daniel Cremers
Abstract: The Laplace-Beltrami operator (LBO) emerges from studying manifolds equipped with a Riemannian metric. It is often called the Swiss army knife of geometry processing as it allows to capture intrinsic shape information and gives rise to heat diffusion, geodesic distances, and a multitude of shape descriptors. It also plays a central role in geometric deep learning. In this work, we explore Finsler manifolds as a generalization of Riemannian manifolds. We revisit the Finsler heat equation and derive a Finsler heat kernel and a Finsler-Laplace-Beltrami Operator (FLBO): a novel theoretically justified anisotropic Laplace-Beltrami operator (ALBO). In experimental evaluations we demonstrate that the proposed FLBO is a valuable alternative to the traditional Riemannian-based LBO and ALBOs for spatial filtering and shape correspondence estimation. We hope that the proposed Finsler heat kernel and the FLBO will inspire further exploration of Finsler geometry in the computer vision community.
Authors: Lili Liang, Guanglu Sun, Jin Qiu, Lizhong Zhang
Abstract: Compositional spatio-temporal reasoning poses a significant challenge in the field of video question answering (VideoQA). Existing approaches struggle to establish effective symbolic reasoning structures, which are crucial for answering compositional spatio-temporal questions. To address this challenge, we propose a neural-symbolic framework called Neural-Symbolic VideoQA (NS-VideoQA), specifically designed for real-world VideoQA tasks. The uniqueness and superiority of NS-VideoQA are two-fold: 1) It proposes a Scene Parser Network (SPN) to transform static-dynamic video scenes into Symbolic Representation (SR), structuralizing persons, objects, relations, and action chronologies. 2) A Symbolic Reasoning Machine (SRM) is designed for top-down question decompositions and bottom-up compositional reasonings. Specifically, a polymorphic program executor is constructed for internally consistent reasoning from SR to the final answer. As a result, Our NS-VideoQA not only improves the compositional spatio-temporal reasoning in real-world VideoQA task, but also enables step-by-step error analysis by tracing the intermediate results. Experimental evaluations on the AGQA Decomp benchmark demonstrate the effectiveness of the proposed NS-VideoQA framework. Empirical studies further confirm that NS-VideoQA exhibits internal consistency in answering compositional questions and significantly improves the capability of spatio-temporal and logical inference for VideoQA tasks.
Authors: Chayanin Tangwiriyasakul, Pedro Borges, Stefano Moriconi, Paul Wright, Yee-Haur Mah, James Teo, Parashkev Nachev, Sebastien Ourselin, M. Jorge Cardoso
Abstract: Stroke is a leading cause of disability and death. Effective treatment decisions require early and informative vascular imaging. 4D perfusion imaging is ideal but rarely available within the first hour after stroke, whereas plain CT and CTA usually are. Hence, we propose a framework to extract a predicted perfusion map (PPM) derived from CT and CTA images. In all eighteen patients, we found significantly high spatial similarity (with average Spearman's correlation = 0.7893) between our predicted perfusion map (PPM) and the T-max map derived from 4D-CTP. Voxelwise correlations between the PPM and National Institutes of Health Stroke Scale (NIHSS) subscores for L/R hand motor, gaze, and language on a large cohort of 2,110 subjects reliably mapped symptoms to expected infarct locations. Therefore our PPM could serve as an alternative for 4D perfusion imaging, if the latter is unavailable, to investigate blood perfusion in the first hours after hospital admission.
Authors: Jiayin Zhu, Linlin Yang, Angela Yao
Abstract: We present InstructHumans, a novel framework for instruction-driven 3D human texture editing. Existing text-based editing methods use Score Distillation Sampling (SDS) to distill guidance from generative models. This work shows that naively using such scores is harmful to editing as they destroy consistency with the source avatar. Instead, we propose an alternate SDS for Editing (SDS-E) that selectively incorporates subterms of SDS across diffusion timesteps. We further enhance SDS-E with spatial smoothness regularization and gradient-based viewpoint sampling to achieve high-quality edits with sharp and high-fidelity detailing. InstructHumans significantly outperforms existing 3D editing methods, consistent with the initial avatar while faithful to the textual instructions. Project page: https://jyzhu.top/instruct-humans .
Authors: Paola Natalia Ca\~nas, Mikel Garc\'ia, Nerea Aranjuelo, Marcos Nieto, Aitor Iglesias, Igor Rodr\'iguez
Abstract: This paper describes the methodology for building a dynamic risk assessment for ADAS (Advanced Driving Assistance Systems) algorithms in parking scenarios, fusing exterior and interior perception for a better understanding of the scene and a more comprehensive risk estimation. This includes the definition of a dynamic risk methodology that depends on the situation from inside and outside the vehicle, the creation of a multi-sensor dataset of risk assessment for ADAS benchmarking purposes, and a Local Dynamic Map (LDM) that fuses data from the exterior and interior of the car to build an LDM-based Dynamic Risk Assessment System (DRAS).
Authors: Xiangyang Zhu, Renrui Zhang, Bowei He, Ziyu Guo, Jiaming Liu, Han Xiao, Chaoyou Fu, Hao Dong, Peng Gao
Abstract: To reduce the reliance on large-scale datasets, recent works in 3D segmentation resort to few-shot learning. Current 3D few-shot segmentation methods first pre-train models on 'seen' classes, and then evaluate their generalization performance on 'unseen' classes. However, the prior pre-training stage not only introduces excessive time overhead but also incurs a significant domain gap on 'unseen' classes. To tackle these issues, we propose a Non-parametric Network for few-shot 3D Segmentation, Seg-NN, and its Parametric variant, Seg-PN. Without training, Seg-NN extracts dense representations by hand-crafted filters and achieves comparable performance to existing parametric models. Due to the elimination of pre-training, Seg-NN can alleviate the domain gap issue and save a substantial amount of time. Based on Seg-NN, Seg-PN only requires training a lightweight QUEry-Support Transferring (QUEST) module, which enhances the interaction between the support set and query set. Experiments suggest that Seg-PN outperforms previous state-of-the-art method by +4.19% and +7.71% mIoU on S3DIS and ScanNet datasets respectively, while reducing training time by -90%, indicating its effectiveness and efficiency.
Authors: Vladan Stojni\'c, Yannis Kalantidis, Giorgos Tolias
Abstract: Vision-Language Models (VLMs) have demonstrated impressive performance on zero-shot classification, i.e. classification when provided merely with a list of class names. In this paper, we tackle the case of zero-shot classification in the presence of unlabeled data. We leverage the graph structure of the unlabeled data and introduce ZLaP, a method based on label propagation (LP) that utilizes geodesic distances for classification. We tailor LP to graphs containing both text and image features and further propose an efficient method for performing inductive inference based on a dual solution and a sparsification step. We perform extensive experiments to evaluate the effectiveness of our method on 14 common datasets and show that ZLaP outperforms the latest related works. Code: https://github.com/vladan-stojnic/ZLaP
Authors: Wenyi Mo, Tianyu Zhang, Yalong Bai, Bing Su, Ji-Rong Wen, Qing Yang
Abstract: Text-to-image generative models, specifically those based on diffusion models like Imagen and Stable Diffusion, have made substantial advancements. Recently, there has been a surge of interest in the delicate refinement of text prompts. Users assign weights or alter the injection time steps of certain words in the text prompts to improve the quality of generated images. However, the success of fine-control prompts depends on the accuracy of the text prompts and the careful selection of weights and time steps, which requires significant manual intervention. To address this, we introduce the \textbf{P}rompt \textbf{A}uto-\textbf{E}diting (PAE) method. Besides refining the original prompts for image generation, we further employ an online reinforcement learning strategy to explore the weights and injection time steps of each word, leading to the dynamic fine-control prompts. The reward function during training encourages the model to consider aesthetic score, semantic consistency, and user preferences. Experimental results demonstrate that our proposed method effectively improves the original prompts, generating visually more appealing images while maintaining semantic alignment. Code is available at https://github.com/Mowenyii/PAE.
Authors: George Retsinas, Panagiotis P. Filntisis, Radek Danecek, Victoria F. Abrevaya, Anastasios Roussos, Timo Bolkart, Petros Maragos
Abstract: While existing methods for 3D face reconstruction from in-the-wild images excel at recovering the overall face shape, they commonly miss subtle, extreme, asymmetric, or rarely observed expressions. We improve upon these methods with SMIRK (Spatial Modeling for Image-based Reconstruction of Kinesics), which faithfully reconstructs expressive 3D faces from images. We identify two key limitations in existing methods: shortcomings in their self-supervised training formulation, and a lack of expression diversity in the training images. For training, most methods employ differentiable rendering to compare a predicted face mesh with the input image, along with a plethora of additional loss functions. This differentiable rendering loss not only has to provide supervision to optimize for 3D face geometry, camera, albedo, and lighting, which is an ill-posed optimization problem, but the domain gap between rendering and input image further hinders the learning process. Instead, SMIRK replaces the differentiable rendering with a neural rendering module that, given the rendered predicted mesh geometry, and sparsely sampled pixels of the input image, generates a face image. As the neural rendering gets color information from sampled image pixels, supervising with neural rendering-based reconstruction loss can focus solely on the geometry. Further, it enables us to generate images of the input identity with varying expressions while training. These are then utilized as input to the reconstruction model and used as supervision with ground truth geometry. This effectively augments the training data and enhances the generalization for diverse expressions. Our qualitative, quantitative and particularly our perceptual evaluations demonstrate that SMIRK achieves the new state-of-the art performance on accurate expression reconstruction. Project webpage: https://georgeretsi.github.io/smirk/.
Authors: Rui Wang, Chuanfu Shen, Manuel J. Marin-Jimenez, George Q. Huang, Shiqi Yu
Abstract: Current gait recognition research mainly focuses on identifying pedestrians captured by the same type of sensor, neglecting the fact that individuals may be captured by different sensors in order to adapt to various environments. A more practical approach should involve cross-modality matching across different sensors. Hence, this paper focuses on investigating the problem of cross-modality gait recognition, with the objective of accurately identifying pedestrians across diverse vision sensors. We present CrossGait inspired by the feature alignment strategy, capable of cross retrieving diverse data modalities. Specifically, we investigate the cross-modality recognition task by initially extracting features within each modality and subsequently aligning these features across modalities. To further enhance the cross-modality performance, we propose a Prototypical Modality-shared Attention Module that learns modality-shared features from two modality-specific features. Additionally, we design a Cross-modality Feature Adapter that transforms the learned modality-specific features into a unified feature space. Extensive experiments conducted on the SUSTech1K dataset demonstrate the effectiveness of CrossGait: (1) it exhibits promising cross-modality ability in retrieving pedestrians across various modalities from different sensors in diverse scenes, and (2) CrossGait not only learns modality-shared features for cross-modality gait recognition but also maintains modality-specific features for single-modality recognition.
Authors: Vishaal Udandarao, Ameya Prabhu, Adhiraj Ghosh, Yash Sharma, Philip H. S. Torr, Adel Bibi, Samuel Albanie, Matthias Bethge
Abstract: Web-crawled pretraining datasets underlie the impressive "zero-shot" evaluation performance of multimodal models, such as CLIP for classification/retrieval and Stable-Diffusion for image generation. However, it is unclear how meaningful the notion of "zero-shot" generalization is for such multimodal models, as it is not known to what extent their pretraining datasets encompass the downstream concepts targeted for during "zero-shot" evaluation. In this work, we ask: How is the performance of multimodal models on downstream concepts influenced by the frequency of these concepts in their pretraining datasets? We comprehensively investigate this question across 34 models and five standard pretraining datasets (CC-3M, CC-12M, YFCC-15M, LAION-400M, LAION-Aesthetics), generating over 300GB of data artifacts. We consistently find that, far from exhibiting "zero-shot" generalization, multimodal models require exponentially more data to achieve linear improvements in downstream "zero-shot" performance, following a sample inefficient log-linear scaling trend. This trend persists even when controlling for sample-level similarity between pretraining and downstream datasets, and testing on purely synthetic data distributions. Furthermore, upon benchmarking models on long-tailed data sampled based on our analysis, we demonstrate that multimodal models across the board perform poorly. We contribute this long-tail test set as the "Let it Wag!" benchmark to further research in this direction. Taken together, our study reveals an exponential need for training data which implies that the key to "zero-shot" generalization capabilities under large-scale training paradigms remains to be found.
Authors: Botao Ren, Botian Xu, Yifan Pu, Jingyi Wang, Zhidong Deng
Abstract: In many image domains, the spatial distribution of objects in a scene exhibits meaningful patterns governed by their semantic relationships. In most modern detection pipelines, however, the detection proposals are processed independently, overlooking the underlying relationships between objects. In this work, we introduce a transformer-based approach to capture these inter-object relationships to refine classification and regression outcomes for detected objects. Building on two-stage detectors, we tokenize the region of interest (RoI) proposals to be processed by a transformer encoder. Specific spatial and geometric relations are incorporated into the attention weights and adaptively modulated and regularized. Experimental results demonstrate that the proposed method achieves consistent performance improvement on three benchmarks including DOTA-v1.0, DOTA-v1.5, and HRSC 2016, especially ranking first on both DOTA-v1.5 and HRSC 2016. Specifically, our new method has an increase of 1.59 mAP on DOTA-v1.0, 4.88 mAP on DOTA-v1.5, and 2.1 mAP on HRSC 2016, respectively, compared to the baselines.
Authors: Junbo Li, Keyan Chen, Gengju Tian, Lu Li, Zhenwei Shi
Abstract: The segmentation and interpretation of the Martian surface play a pivotal role in Mars exploration, providing essential data for the trajectory planning and obstacle avoidance of rovers. However, the complex topography, similar surface features, and the lack of extensive annotated data pose significant challenges to the high-precision semantic segmentation of the Martian surface. To address these challenges, we propose a novel encoder-decoder based Mars segmentation network, termed MarsSeg. Specifically, we employ an encoder-decoder structure with a minimized number of down-sampling layers to preserve local details. To facilitate a high-level semantic understanding across the shadow multi-level feature maps, we introduce a feature enhancement connection layer situated between the encoder and decoder. This layer incorporates Mini Atrous Spatial Pyramid Pooling (Mini-ASPP), Polarized Self-Attention (PSA), and Strip Pyramid Pooling Module (SPPM). The Mini-ASPP and PSA are specifically designed for shadow feature enhancement, thereby enabling the expression of local details and small objects. Conversely, the SPPM is employed for deep feature enhancement, facilitating the extraction of high-level semantic category-related information. Experimental results derived from the Mars-Seg and AI4Mars datasets substantiate that the proposed MarsSeg outperforms other state-of-the-art methods in segmentation performance, validating the efficacy of each proposed component.
Authors: Mengting Li, Chuang Zhu
Abstract: In recent years, deep neural networks (DNNs) have gained remarkable achievement in computer vision tasks, and the success of DNNs often depends greatly on the richness of data. However, the acquisition process of data and high-quality ground truth requires a lot of manpower and money. In the long, tedious process of data annotation, annotators are prone to make mistakes, resulting in incorrect labels of images, i.e., noisy labels. The emergence of noisy labels is inevitable. Moreover, since research shows that DNNs can easily fit noisy labels, the existence of noisy labels will cause significant damage to the model training process. Therefore, it is crucial to combat noisy labels for computer vision tasks, especially for classification tasks. In this survey, we first comprehensively review the evolution of different deep learning approaches for noisy label combating in the image classification task. In addition, we also review different noise patterns that have been proposed to design robust algorithms. Furthermore, we explore the inner pattern of real-world label noise and propose an algorithm to generate a synthetic label noise pattern guided by real-world data. We test the algorithm on the well-known real-world dataset CIFAR-10N to form a new real-world data-guided synthetic benchmark and evaluate some typical noise-robust methods on the benchmark.
Authors: Weile Li, Muqing Shi, Zhonghua Hong
Abstract: Traditional deep learning-based object detection networks often resize images during the data preprocessing stage to achieve a uniform size and scale in the feature map. Resizing is done to facilitate model propagation and fully connected classification. However, resizing inevitably leads to object deformation and loss of valuable information in the images. This drawback becomes particularly pronounced for tiny objects like distribution towers with linear shapes and few pixels. To address this issue, we propose abandoning the resizing operation. Instead, we introduce Positional-Encoding Multi-head Criss-Cross Attention. This allows the model to capture contextual information and learn from multiple representation subspaces, effectively enriching the semantics of distribution towers. Additionally, we enhance Spatial Pyramid Pooling by reshaping three pooled feature maps into a new unified one while also reducing the computational burden. This approach allows images of different sizes and scales to generate feature maps with uniform dimensions and can be employed in feature map propagation. Our SCAResNet incorporates these aforementioned improvements into the backbone network ResNet. We evaluated our SCAResNet using the Electric Transmission and Distribution Infrastructure Imagery dataset from Duke University. Without any additional tricks, we employed various object detection models with Gaussian Receptive Field based Label Assignment as the baseline. When incorporating the SCAResNet into the baseline model, we achieved a 2.1% improvement in mAPs. This demonstrates the advantages of our SCAResNet in detecting transmission and distribution towers and its value in tiny object detection. The source code is available at https://github.com/LisavilaLee/SCAResNet_mmdet.
Authors: Fran\c{c}ois Darmon, Lorenzo Porzi, Samuel Rota-Bul\`o, Peter Kontschieder
Abstract: In this paper, we address common error sources for 3D Gaussian Splatting (3DGS) including blur, imperfect camera poses, and color inconsistencies, with the goal of improving its robustness for practical applications like reconstructions from handheld phone captures. Our main contribution involves modeling motion blur as a Gaussian distribution over camera poses, allowing us to address both camera pose refinement and motion blur correction in a unified way. Additionally, we propose mechanisms for defocus blur compensation and for addressing color in-consistencies caused by ambient light, shadows, or due to camera-related factors like varying white balancing settings. Our proposed solutions integrate in a seamless way with the 3DGS formulation while maintaining its benefits in terms of training efficiency and rendering speed. We experimentally validate our contributions on relevant benchmark datasets including Scannet++ and Deblur-NeRF, obtaining state-of-the-art results and thus consistent improvements over relevant baselines.
Authors: Ji-Jia Wu, Andy Chia-Hao Chang, Chieh-Yu Chuang, Chun-Pei Chen, Yu-Lun Liu, Min-Hung Chen, Hou-Ning Hu, Yung-Yu Chuang, Yen-Yu Lin
Abstract: This paper addresses text-supervised semantic segmentation, aiming to learn a model capable of segmenting arbitrary visual concepts within images by using only image-text pairs without dense annotations. Existing methods have demonstrated that contrastive learning on image-text pairs effectively aligns visual segments with the meanings of texts. We notice that there is a discrepancy between text alignment and semantic segmentation: A text often consists of multiple semantic concepts, whereas semantic segmentation strives to create semantically homogeneous segments. To address this issue, we propose a novel framework, Image-Text Co-Decomposition (CoDe), where the paired image and text are jointly decomposed into a set of image regions and a set of word segments, respectively, and contrastive learning is developed to enforce region-word alignment. To work with a vision-language model, we present a prompt learning mechanism that derives an extra representation to highlight an image segment or a word segment of interest, with which more effective features can be extracted from that segment. Comprehensive experimental results demonstrate that our method performs favorably against existing text-supervised semantic segmentation methods on six benchmark datasets.
Authors: Albert J. Zhai, Yuan Shen, Emily Y. Chen, Gloria X. Wang, Xinlei Wang, Sheng Wang, Kaiyu Guan, Shenlong Wang
Abstract: Can computers perceive the physical properties of objects solely through vision? Research in cognitive science and vision science has shown that humans excel at identifying materials and estimating their physical properties based purely on visual appearance. In this paper, we present a novel approach for dense prediction of the physical properties of objects using a collection of images. Inspired by how humans reason about physics through vision, we leverage large language models to propose candidate materials for each object. We then construct a language-embedded point cloud and estimate the physical properties of each 3D point using a zero-shot kernel regression approach. Our method is accurate, annotation-free, and applicable to any object in the open world. Experiments demonstrate the effectiveness of the proposed approach in various physical property reasoning tasks, such as estimating the mass of common objects, as well as other properties like friction and hardness.
Authors: Sangwon Jang, Jaehyeong Jo, Kimin Lee, Sung Ju Hwang
Abstract: Text-to-image diffusion models have shown remarkable success in generating a personalized subject based on a few reference images. However, current methods struggle with handling multiple subjects simultaneously, often resulting in mixed identities with combined attributes from different subjects. In this work, we present MuDI, a novel framework that enables multi-subject personalization by effectively decoupling identities from multiple subjects. Our main idea is to utilize segmented subjects generated by the Segment Anything Model for both training and inference, as a form of data augmentation for training and initialization for the generation process. Our experiments demonstrate that MuDI can produce high-quality personalized images without identity mixing, even for highly similar subjects as shown in Figure 1. In human evaluation, MuDI shows twice as many successes for personalizing multiple subjects without identity mixing over existing baselines and is preferred over 70% compared to the strongest baseline. More results are available at https://mudi-t2i.github.io/.
Authors: Jiong Wu
Abstract: Existing unsupervised deformable image registration methods usually rely on metrics applied to the gradients of predicted displacement or velocity fields as a regularization term to ensure transformation smoothness, which potentially limits registration accuracy. In this study, we propose a novel approach to enhance unsupervised deformable image registration by introducing a new differential operator into the registration framework. This operator, acting on the velocity field and mapping it to a dual space, ensures the smoothness of the velocity field during optimization, facilitating accurate deformable registration. In addition, to tackle the challenge of capturing large deformations inside image pairs, we introduce a Cross-Coordinate Attention module (CCA) and embed it into a proposed Fully Convolutional Networks (FCNs)-based multi-resolution registration architecture. Evaluation experiments are conducted on two magnetic resonance imaging (MRI) datasets. Compared to various state-of-the-art registration approaches, including a traditional algorithm and three representative unsupervised learning-based methods, our method achieves superior accuracies, maintaining desirable diffeomorphic properties, and exhibiting promising registration speed.
Authors: Michael Saxon, Fatima Jahara, Mahsa Khoshnoodi, Yujie Lu, Aditya Sharma, William Yang Wang
Abstract: With advances in the quality of text-to-image (T2I) models has come interest in benchmarking their prompt faithfulness-the semantic coherence of generated images to the prompts they were conditioned on. A variety of T2I faithfulness metrics have been proposed, leveraging advances in cross-modal embeddings and vision-language models (VLMs). However, these metrics are not rigorously compared and benchmarked, instead presented against few weak baselines by correlation to human Likert scores over a set of easy-to-discriminate images. We introduce T2IScoreScore (TS2), a curated set of semantic error graphs containing a prompt and a set increasingly erroneous images. These allow us to rigorously judge whether a given prompt faithfulness metric can correctly order images with respect to their objective error count and significantly discriminate between different error nodes, using meta-metric scores derived from established statistical tests. Surprisingly, we find that the state-of-the-art VLM-based metrics (e.g., TIFA, DSG, LLMScore, VIEScore) we tested fail to significantly outperform simple feature-based metrics like CLIPScore, particularly on a hard subset of naturally-occurring T2I model errors. TS2 will enable the development of better T2I prompt faithfulness metrics through more rigorous comparison of their conformity to expected orderings and separations under objective criteria.
Authors: Zifu Wan, Yuhao Wang, Silong Yong, Pingping Zhang, Simon Stepputtis, Katia Sycara, Yaqi Xie
Abstract: Multi-modal semantic segmentation significantly enhances AI agents' perception and scene understanding, especially under adverse conditions like low-light or overexposed environments. Leveraging additional modalities (X-modality) like thermal and depth alongside traditional RGB provides complementary information, enabling more robust and reliable segmentation. In this work, we introduce Sigma, a Siamese Mamba network for multi-modal semantic segmentation, utilizing the Selective Structured State Space Model, Mamba. Unlike conventional methods that rely on CNNs, with their limited local receptive fields, or Vision Transformers (ViTs), which offer global receptive fields at the cost of quadratic complexity, our model achieves global receptive fields coverage with linear complexity. By employing a Siamese encoder and innovating a Mamba fusion mechanism, we effectively select essential information from different modalities. A decoder is then developed to enhance the channel-wise modeling ability of the model. Our method, Sigma, is rigorously evaluated on both RGB-Thermal and RGB-Depth segmentation tasks, demonstrating its superiority and marking the first successful application of State Space Models (SSMs) in multi-modal perception tasks. Code is available at https://github.com/zifuwan/Sigma.
Authors: Man Yao, Jiakui Hu, Tianxiang Hu, Yifan Xu, Zhaokun Zhou, Yonghong Tian, Bo Xu, Guoqi Li
Abstract: Neuromorphic computing, which exploits Spiking Neural Networks (SNNs) on neuromorphic chips, is a promising energy-efficient alternative to traditional AI. CNN-based SNNs are the current mainstream of neuromorphic computing. By contrast, no neuromorphic chips are designed especially for Transformer-based SNNs, which have just emerged, and their performance is only on par with CNN-based SNNs, offering no distinct advantage. In this work, we propose a general Transformer-based SNN architecture, termed as ``Meta-SpikeFormer", whose goals are: 1) Lower-power, supports the spike-driven paradigm that there is only sparse addition in the network; 2) Versatility, handles various vision tasks; 3) High-performance, shows overwhelming performance advantages over CNN-based SNNs; 4) Meta-architecture, provides inspiration for future next-generation Transformer-based neuromorphic chip designs. Specifically, we extend the Spike-driven Transformer in \citet{yao2023spike} into a meta architecture, and explore the impact of structure, spike-driven self-attention, and skip connection on its performance. On ImageNet-1K, Meta-SpikeFormer achieves 80.0\% top-1 accuracy (55M), surpassing the current state-of-the-art (SOTA) SNN baselines (66M) by 3.7\%. This is the first direct training SNN backbone that can simultaneously supports classification, detection, and segmentation, obtaining SOTA results in SNNs. Finally, we discuss the inspiration of the meta SNN architecture for neuromorphic chip design. Source code and models are available at \url{https://github.com/BICLab/Spike-Driven-Transformer-V2}.
Authors: Dhananjay Saikumar, Blesson Varghese
Abstract: Modern deep neural networks (DNNs) consist of millions of parameters, necessitating high-performance computing during training and inference. Pruning is one solution that significantly reduces the space and time complexities of DNNs. Traditional pruning methods that are applied post-training focus on streamlining inference, but there are recent efforts to leverage sparsity early on by pruning before training. Pruning methods, such as iterative magnitude-based pruning (IMP) achieve up to a 90% parameter reduction while retaining accuracy comparable to the original model. However, this leads to impractical runtime as it relies on multiple train-prune-reset cycles to identify and eliminate redundant parameters. In contrast, training agnostic early pruning methods, such as SNIP and SynFlow offer fast pruning but fall short of the accuracy achieved by IMP at high sparsities. To bridge this gap, we present Dual Gradient-Based Rapid Iterative Pruning (DRIVE), which leverages dense training for initial epochs to counteract the randomness inherent at the initialization. Subsequently, it employs a unique dual gradient-based metric for parameter ranking. It has been experimentally demonstrated for VGG and ResNet architectures on CIFAR-10/100 and Tiny ImageNet, and ResNet on ImageNet that DRIVE consistently has superior performance over other training-agnostic early pruning methods in accuracy. Notably, DRIVE is 43$\times$ to 869$\times$ faster than IMP for pruning.
Authors: Elodie Germani (EMPENN, LACODAM), Elisa Fromont (LACODAM), Camille Maumet (EMPENN)
Abstract: We propose a novel approach to improve the reproducibility of neuroimaging results by converting statistic maps across different functional MRI pipelines. We make the assumption that pipelines can be considered as a style component of data and propose to use different generative models, among which, Diffusion Models (DM) to convert data between pipelines. We design a new DM-based unsupervised multi-domain image-to-image transition framework and constrain the generation of 3D fMRI statistic maps using the latent space of an auxiliary classifier that distinguishes statistic maps from different pipelines. We extend traditional sampling techniques used in DM to improve the transition performance. Our experiments demonstrate that our proposed methods are successful: pipelines can indeed be transferred, providing an important source of data augmentation for future medical studies.
Authors: Angus Nicolson, Lisa Schut, J. Alison Noble, Yarin Gal
Abstract: Recent interpretability methods propose using concept-based explanations to translate the internal representations of deep learning models into a language that humans are familiar with: concepts. This requires understanding which concepts are present in the representation space of a neural network. One popular method for finding concepts is Concept Activation Vectors (CAVs), which are learnt using a probe dataset of concept exemplars. In this work, we investigate three properties of CAVs. CAVs may be: (1) inconsistent between layers, (2) entangled with different concepts, and (3) spatially dependent. Each property provides both challenges and opportunities in interpreting models. We introduce tools designed to detect the presence of these properties, provide insight into how they affect the derived explanations, and provide recommendations to minimise their impact. Understanding these properties can be used to our advantage. For example, we introduce spatially dependent CAVs to test if a model is translation invariant with respect to a specific concept and class. Our experiments are performed on ImageNet and a new synthetic dataset, Elements. Elements is designed to capture a known ground truth relationship between concepts and classes. We release this dataset to facilitate further research in understanding and evaluating interpretability methods.
Authors: Afonso Oliveira, Nuno Fachada, Jo\~ao P. Matos-Carvalho
Abstract: The integration of data science into Geographic Information Systems (GIS) has facilitated the evolution of these tools into complete spatial analysis platforms. The adoption of machine learning and big data techniques has equipped these platforms with the capacity to handle larger amounts of increasingly complex data, transcending the limitations of more traditional approaches. This work traces the historical and technical evolution of data science and GIS as fields of study, highlighting the critical points of convergence between domains, and underlining the many sectors that rely on this integration. A GIS application is presented as a case study in the disaster management sector where we utilize aerial data from Tr\'oia, Portugal, to emphasize the process of insight extraction from raw data. We conclude by outlining prospects for future research in integration of these fields in general, and the developed application in particular.
Authors: Sabyasachi Sahoo, Mostafa ElAraby, Jonas Ngnawe, Yann Pequignot, Frederic Precioso, Christian Gagne
Abstract: Test Time Adaptation (TTA) addresses the problem of distribution shift by enabling pretrained models to learn new features on an unseen domain at test time. However, it poses a significant challenge to maintain a balance between learning new features and retaining useful pretrained features. In this paper, we propose Layerwise EArly STopping (LEAST) for TTA to address this problem. The key idea is to stop adapting individual layers during TTA if the features being learned do not appear beneficial for the new domain. For that purpose, we propose using a novel gradient-based metric to measure the relevance of the current learnt features to the new domain without the need for supervised labels. More specifically, we propose to use this metric to determine dynamically when to stop updating each layer during TTA. This enables a more balanced adaptation, restricted to layers benefiting from it, and only for a certain number of steps. Such an approach also has the added effect of limiting the forgetting of pretrained features useful for dealing with new domains. Through extensive experiments, we demonstrate that Layerwise Early Stopping improves the performance of existing TTA approaches across multiple datasets, domain shifts, model architectures, and TTA losses.
Authors: Zitao Shuai, Liyue Shen
Abstract: Vision-language pre-training (VLP) has arised as an efficient scheme for multimodal representation learning, but it requires large-scale multimodal data for pre-training, making it an obstacle especially for biomedical applications. To overcome the data limitation, federated learning (FL) can be a promising strategy to scale up the dataset for biomedical VLP while protecting data privacy. However, client data are often heterogeneous in real-world scenarios, and we observe that local training on heterogeneous client data would distort the multimodal representation learning and lead to biased cross-modal alignment. To address this challenge, we propose Federated distributional Robust Guidance-Based (FedRGB) learning framework for federated VLP with robustness to data heterogeneity. Specifically, we utilize a guidance-based local training scheme to reduce feature distortions, and employ a distribution-based min-max optimization to learn unbiased cross-modal alignment. The experiments on real-world datasets show our method successfully promotes efficient federated multimodal learning for biomedical VLP with data heterogeneity.
Authors: Judy X Yang, Jun Zhou, Jing Wang, Hui Tian, Wee Chung Liew
Abstract: The fusion of hyperspectral and LiDAR data has been an active research topic. Existing fusion methods have ignored the high-dimensionality and redundancy challenges in hyperspectral images, despite that band selection methods have been intensively studied for hyperspectral image (HSI) processing. This paper addresses this significant gap by introducing a cross-attention mechanism from the transformer architecture for the selection of HSI bands guided by LiDAR data. LiDAR provides high-resolution vertical structural information, which can be useful in distinguishing different types of land cover that may have similar spectral signatures but different structural profiles. In our approach, the LiDAR data are used as the "query" to search and identify the "key" from the HSI to choose the most pertinent bands for LiDAR. This method ensures that the selected HSI bands drastically reduce redundancy and computational requirements while working optimally with the LiDAR data. Extensive experiments have been undertaken on three paired HSI and LiDAR data sets: Houston 2013, Trento and MUUFL. The results highlight the superiority of the cross-attention mechanism, underlining the enhanced classification accuracy of the identified HSI bands when fused with the LiDAR features. The results also show that the use of fewer bands combined with LiDAR surpasses the performance of state-of-the-art fusion models.
Authors: Nimrod Shabtay, Eli Schwartz, Raja Giryes
Abstract: Phase-coded imaging is a computational imaging method designed to tackle tasks such as passive depth estimation and extended depth of field (EDOF) using depth cues inserted during image capture. Most of the current deep learning-based methods for depth estimation or all-in-focus imaging require a training dataset with high-quality depth maps and an optimal focus point at infinity for all-in-focus images. Such datasets are difficult to create, usually synthetic, and require external graphic programs. We propose a new method named "Deep Phase Coded Image Prior" (DPCIP) for jointly recovering the depth map and all-in-focus image from a coded-phase image using solely the captured image and the optical information of the imaging system. Our approach does not depend on any specific dataset and surpasses prior supervised techniques utilizing the same imaging system. This improvement is achieved through the utilization of a problem formulation based on implicit neural representation (INR) and deep image prior (DIP). Due to our zero-shot method, we overcome the barrier of acquiring accurate ground-truth data of depth maps and all-in-focus images for each new phase-coded system introduced. This allows focusing mainly on developing the imaging system, and not on ground-truth data collection.
Authors: Daniel Panangian, Ksenia Bittner
Abstract: A low-resolution digital surface model (DSM) features distinctive attributes impacted by noise, sensor limitations and data acquisition conditions, which failed to be replicated using simple interpolation methods like bicubic. This causes super-resolution models trained on synthetic data does not perform effectively on real ones. Training a model on real low and high resolution DSMs pairs is also a challenge because of the lack of information. On the other hand, the existence of other imaging modalities of the same scene can be used to enrich the information needed for large-scale super-resolution. In this work, we introduce a novel methodology to address the intricacies of real-world DSM super-resolution, named REAL-GDSR, breaking down this ill-posed problem into two steps. The first step involves the utilization of a residual local refinement network. This strategic approach departs from conventional methods that trained to directly predict height values instead of the differences (residuals) and utilize large receptive fields in their networks. The second step introduces a diffusion-based technique that enhances the results on a global scale, with a primary focus on smoothing and edge preservation. Our experiments underscore the effectiveness of the proposed method. We conduct a comprehensive evaluation, comparing it to recent state-of-the-art techniques in the domain of real-world DSM super-resolution (SR). Our approach consistently outperforms these existing methods, as evidenced through qualitative and quantitative assessments.
Authors: Shahzad Ali, Yu Rim Lee, Soo Young Park, Won Young Tak, Soon Ki Jung
Abstract: Downsampling images and labels, often necessitated by limited resources or to expedite network training, leads to the loss of small objects and thin boundaries. This undermines the segmentation network's capacity to interpret images accurately and predict detailed labels, resulting in diminished performance compared to processing at original resolutions. This situation exemplifies the trade-off between efficiency and accuracy, with higher downsampling factors further impairing segmentation outcomes. Preserving information during downsampling is especially critical for medical image segmentation tasks. To tackle this challenge, we introduce a novel method named Edge-preserving Probabilistic Downsampling (EPD). It utilizes class uncertainty within a local window to produce soft labels, with the window size dictating the downsampling factor. This enables a network to produce quality predictions at low resolutions. Beyond preserving edge details more effectively than conventional nearest-neighbor downsampling, employing a similar algorithm for images, it surpasses bilinear interpolation in image downsampling, enhancing overall performance. Our method significantly improved Intersection over Union (IoU) to 2.85%, 8.65%, and 11.89% when downsampling data to 1/2, 1/4, and 1/8, respectively, compared to conventional interpolation methods.
Authors: Mohammed Ghaith Altarabichi, S{\l}awomir Nowaczyk, Sepideh Pashami, Peyman Sheikholharam Mashhadi, Julia Handl
Abstract: This paper investigates how various randomization techniques impact Deep Neural Networks (DNNs). Randomization, like weight noise and dropout, aids in reducing overfitting and enhancing generalization, but their interactions are poorly understood. The study categorizes randomness techniques into four types and proposes new methods: adding noise to the loss function and random masking of gradient updates. Using Particle Swarm Optimizer (PSO) for hyperparameter optimization, it explores optimal configurations across MNIST, FASHION-MNIST, CIFAR10, and CIFAR100 datasets. Over 30,000 configurations are evaluated, revealing data augmentation and weight initialization randomness as main performance contributors. Correlation analysis shows different optimizers prefer distinct randomization types. The complete implementation and dataset are available on GitHub.
Authors: Chenyang Wu, Yifan Duan, Xinran Zhang, Yu Sheng, Jianmin Ji, Yanyong Zhang
Abstract: Localization and mapping are critical tasks for various applications such as autonomous vehicles and robotics. The challenges posed by outdoor environments present particular complexities due to their unbounded characteristics. In this work, we present MM-Gaussian, a LiDAR-camera multi-modal fusion system for localization and mapping in unbounded scenes. Our approach is inspired by the recently developed 3D Gaussians, which demonstrate remarkable capabilities in achieving high rendering quality and fast rendering speed. Specifically, our system fully utilizes the geometric structure information provided by solid-state LiDAR to address the problem of inaccurate depth encountered when relying solely on visual solutions in unbounded, outdoor scenarios. Additionally, we utilize 3D Gaussian point clouds, with the assistance of pixel-level gradient descent, to fully exploit the color information in photos, thereby achieving realistic rendering effects. To further bolster the robustness of our system, we designed a relocalization module, which assists in returning to the correct trajectory in the event of a localization failure. Experiments conducted in multiple scenarios demonstrate the effectiveness of our method.
Authors: Mingyuan Zhou, Huangjie Zheng, Zhendong Wang, Mingzhang Yin, Hai Huang
Abstract: We introduce Score identity Distillation (SiD), an innovative data-free method that distills the generative capabilities of pretrained diffusion models into a single-step generator. SiD not only facilitates an exponentially fast reduction in Fr\'echet inception distance (FID) during distillation but also approaches or even exceeds the FID performance of the original teacher diffusion models. By reformulating forward diffusion processes as semi-implicit distributions, we leverage three score-related identities to create an innovative loss mechanism. This mechanism achieves rapid FID reduction by training the generator using its own synthesized images, eliminating the need for real data or reverse-diffusion-based generation, all accomplished within significantly shortened generation time. Upon evaluation across four benchmark datasets, the SiD algorithm demonstrates high iteration efficiency during distillation and surpasses competing distillation approaches, whether they are one-step or few-step, data-free, or dependent on training data, in terms of generation quality. This achievement not only redefines the benchmarks for efficiency and effectiveness in diffusion distillation but also in the broader field of diffusion-based generation. Our PyTorch implementation will be publicly accessible on GitHub.
Authors: Jianxin Zhou, Kadishe Fejza, Massimiliano Salvatori, Daniele Della Latta, Gregory M. Hermann, Angela Di Fulvio
Abstract: Our understanding of organs at risk is progressing to include physical small tissues such as coronary arteries and the radiosensitivities of many small organs and tissues are high. Therefore, the accurate segmentation of small volumes in external radiotherapy is crucial to protect them from over-irradiation. Moreover, with the development of the particle therapy and on-board imaging, the treatment becomes more accurate and precise. The purpose of this work is to optimize organ segmentation algorithms for small organs. We used 50 three-dimensional (3-D) computed tomography (CT) head and neck images from StructSeg2019 challenge to develop a general-purpose V-Net model to segment 20 organs in the head and neck region. We applied specific strategies to improve the segmentation accuracy of the small volumes in this anatomical region, i.e., the lens of the eye. Then, we used 17 additional head images from OSF healthcare to validate the robustness of the V Net model optimized for small-volume segmentation. With the study of the StructSeg2019 images, we found that the optimization of the image normalization range and classification threshold yielded a segmentation improvement of the lens of the eye of approximately 50%, compared to the use of the V-Net not optimized for small volumes. We used the optimized model to segment 17 images acquired using heterogeneous protocols. We obtained comparable Dice coefficient values for the clinical and StructSeg2019 images (0.61 plus/minus 0.07 and 0.58 plus/minus 0.10 for the left and right lens of the eye, respectively)
Authors: Trilokesh Ranjan Sarkar, Nilanjan Das, Pralay Sankar Maitra, Bijoy Some, Ritwik Saha, Orijita Adhikary, Bishal Bose, Jaydip Sen
Abstract: This technical report delves into an in-depth exploration of adversarial attacks specifically targeted at Deep Neural Networks (DNNs) utilized for image classification. The study also investigates defense mechanisms aimed at bolstering the robustness of machine learning models. The research focuses on comprehending the ramifications of two prominent attack methodologies: the Fast Gradient Sign Method (FGSM) and the Carlini-Wagner (CW) approach. These attacks are examined concerning three pre-trained image classifiers: Resnext50_32x4d, DenseNet-201, and VGG-19, utilizing the Tiny-ImageNet dataset. Furthermore, the study proposes the robustness of defensive distillation as a defense mechanism to counter FGSM and CW attacks. This defense mechanism is evaluated using the CIFAR-10 dataset, where CNN models, specifically resnet101 and Resnext50_32x4d, serve as the teacher and student models, respectively. The proposed defensive distillation model exhibits effectiveness in thwarting attacks such as FGSM. However, it is noted to remain susceptible to more sophisticated techniques like the CW attack. The document presents a meticulous validation of the proposed scheme. It provides detailed and comprehensive results, elucidating the efficacy and limitations of the defense mechanisms employed. Through rigorous experimentation and analysis, the study offers insights into the dynamics of adversarial attacks on DNNs, as well as the effectiveness of defensive strategies in mitigating their impact.
Authors: Zhengyuan Jiang, Moyang Guo, Yuepeng Hu, Neil Zhenqiang Gong
Abstract: Several companies--such as Google, Microsoft, and OpenAI--have deployed techniques to watermark AI-generated content to enable proactive detection. However, existing literature mainly focuses on user-agnostic detection. Attribution aims to further trace back the user of a generative-AI service who generated a given content detected as AI-generated. Despite its growing importance, attribution is largely unexplored. In this work, we aim to bridge this gap by providing the first systematic study on watermark-based, user-aware detection and attribution of AI-generated content. Specifically, we theoretically study the detection and attribution performance via rigorous probabilistic analysis. Moreover, we develop an efficient algorithm to select watermarks for the users to enhance attribution performance. Both our theoretical and empirical results show that watermark-based detection and attribution inherit the accuracy and (non-)robustness properties of the watermarking method.
Authors: Alexander Kroner, Mario Senden, Kurt Driessens, Rainer Goebel
Abstract: Predicting salient regions in natural images requires the detection of objects that are present in a scene. To develop robust representations for this challenging task, high-level visual features at multiple spatial scales must be extracted and augmented with contextual information. However, existing models aimed at explaining human fixation maps do not incorporate such a mechanism explicitly. Here we propose an approach based on a convolutional neural network pre-trained on a large-scale image classification task. The architecture forms an encoder-decoder structure and includes a module with multiple convolutional layers at different dilation rates to capture multi-scale features in parallel. Moreover, we combine the resulting representations with global scene information for accurately predicting visual saliency. Our model achieves competitive and consistent results across multiple evaluation metrics on two public saliency benchmarks and we demonstrate the effectiveness of the suggested approach on five datasets and selected examples. Compared to state of the art approaches, the network is based on a lightweight image classification backbone and hence presents a suitable choice for applications with limited computational resources, such as (virtual) robotic systems, to estimate human fixations across complex natural scenes.
Authors: Przemys{\l}aw Spurek, Sebastian Winczowski, Maciej Zi\k{e}ba, Tomasz Trzci\'nski, Kacper Kania, Marcin Mazur
Abstract: Recently proposed 3D object reconstruction methods represent a mesh with an atlas - a set of planar patches approximating the surface. However, their application in a real-world scenario is limited since the surfaces of reconstructed objects contain discontinuities, which degrades the quality of the final mesh. This is mainly caused by independent processing of individual patches, and in this work, we postulate to mitigate this limitation by preserving local consistency around patch vertices. To that end, we introduce a Locally Conditioned Atlas (LoCondA), a framework for representing a 3D object hierarchically in a generative model. Firstly, the model maps a point cloud of an object into a sphere. Secondly, by leveraging a spherical prior, we enforce the mapping to be locally consistent on the sphere and on the target object. This way, we can sample a mesh quad on that sphere and project it back onto the object's manifold. With LoCondA, we can produce topologically diverse objects while maintaining quads to be stitched together. We show that the proposed approach provides structurally coherent reconstructions while producing meshes of quality comparable to the competitors.
Authors: Hanwei Zhang, Felipe Torres, Ronan Sicre, Yannis Avrithis, Stephane Ayache
Abstract: Methods based on class activation maps (CAM) provide a simple mechanism to interpret predictions of convolutional neural networks by using linear combinations of feature maps as saliency maps. By contrast, masking-based methods optimize a saliency map directly in the image space or learn it by training another network on additional data. In this work we introduce Opti-CAM, combining ideas from CAM-based and masking-based approaches. Our saliency map is a linear combination of feature maps, where weights are optimized per image such that the logit of the masked image for a given class is maximized. We also fix a fundamental flaw in two of the most common evaluation metrics of attribution methods. On several datasets, Opti-CAM largely outperforms other CAM-based approaches according to the most relevant classification metrics. We provide empirical evidence supporting that localization and classifier interpretability are not necessarily aligned.
Authors: Ntumba Elie Nsampi, Adarsh Djeacoumar, Hans-Peter Seidel, Tobias Ritschel, Thomas Leimk\"uhler
Abstract: Neural fields are evolving towards a general-purpose continuous representation for visual computing. Yet, despite their numerous appealing properties, they are hardly amenable to signal processing. As a remedy, we present a method to perform general continuous convolutions with general continuous signals such as neural fields. Observing that piecewise polynomial kernels reduce to a sparse set of Dirac deltas after repeated differentiation, we leverage convolution identities and train a repeated integral field to efficiently execute large-scale convolutions. We demonstrate our approach on a variety of data modalities and spatially-varying kernels.
Authors: Antyanta Bangunharcana, Ahmed Magd, Kyung-Soo Kim
Abstract: Self-supervised multi-frame depth estimation achieves high accuracy by computing matching costs of pixel correspondences between adjacent frames, injecting geometric information into the network. These pixel-correspondence candidates are computed based on the relative pose estimates between the frames. Accurate pose predictions are essential for precise matching cost computation as they influence the epipolar geometry. Furthermore, improved depth estimates can, in turn, be used to align pose estimates. Inspired by traditional structure-from-motion (SfM) principles, we propose the DualRefine model, which tightly couples depth and pose estimation through a feedback loop. Our novel update pipeline uses a deep equilibrium model framework to iteratively refine depth estimates and a hidden state of feature maps by computing local matching costs based on epipolar geometry. Importantly, we used the refined depth estimates and feature maps to compute pose updates at each step. This update in the pose estimates slowly alters the epipolar geometry during the refinement process. Experimental results on the KITTI dataset demonstrate competitive depth prediction and odometry prediction performance surpassing published self-supervised baselines.
Authors: Burak Ercan, Onur Eker, Aykut Erdem, Erkut Erdem
Abstract: Event cameras are a new type of vision sensor that incorporates asynchronous and independent pixels, offering advantages over traditional frame-based cameras such as high dynamic range and minimal motion blur. However, their output is not easily understandable by humans, making the reconstruction of intensity images from event streams a fundamental task in event-based vision. While recent deep learning-based methods have shown promise in video reconstruction from events, this problem is not completely solved yet. To facilitate comparison between different approaches, standardized evaluation protocols and diverse test datasets are essential. This paper proposes a unified evaluation methodology and introduces an open-source framework called EVREAL to comprehensively benchmark and analyze various event-based video reconstruction methods from the literature. Using EVREAL, we give a detailed analysis of the state-of-the-art methods for event-based video reconstruction, and provide valuable insights into the performance of these methods under varying settings, challenging scenarios, and downstream tasks.
Authors: Tan Wang, Linjie Li, Kevin Lin, Yuanhao Zhai, Chung-Ching Lin, Zhengyuan Yang, Hanwang Zhang, Zicheng Liu, Lijuan Wang
Abstract: Generative AI has made significant strides in computer vision, particularly in text-driven image/video synthesis (T2I/T2V). Despite the notable advancements, it remains challenging in human-centric content synthesis such as realistic dance generation. Current methodologies, primarily tailored for human motion transfer, encounter difficulties when confronted with real-world dance scenarios (e.g., social media dance), which require to generalize across a wide spectrum of poses and intricate human details. In this paper, we depart from the traditional paradigm of human motion transfer and emphasize two additional critical attributes for the synthesis of human dance content in social media contexts: (i) Generalizability: the model should be able to generalize beyond generic human viewpoints as well as unseen human subjects, backgrounds, and poses; (ii) Compositionality: it should allow for the seamless composition of seen/unseen subjects, backgrounds, and poses from different sources. To address these challenges, we introduce DISCO, which includes a novel model architecture with disentangled control to improve the compositionality of dance synthesis, and an effective human attribute pre-training for better generalizability to unseen humans. Extensive qualitative and quantitative results demonstrate that DisCc can generate high-quality human dance images and videos with diverse appearances and flexible motions. Code is available at https://disco-dance.github.io/.
Authors: Zijie Song, Zhenzhen Hu, Yuanen Zhou, Ye Zhao, Richang Hong, Meng Wang
Abstract: Cross-lingual image captioning is a challenging task that requires addressing both cross-lingual and cross-modal obstacles in multimedia analysis. The crucial issue in this task is to model the global and the local matching between the image and different languages. Existing cross-modal embedding methods based on the transformer architecture oversee the local matching between the image region and monolingual words, especially when dealing with diverse languages. To overcome these limitations, we propose an Embedded Heterogeneous Attention Transformer (EHAT) to establish cross-domain relationships and local correspondences between images and different languages by using a heterogeneous network. EHAT comprises Masked Heterogeneous Cross-attention (MHCA), Heterogeneous Attention Reasoning Network (HARN), and Heterogeneous Co-attention (HCA). The HARN serves as the core network and it captures cross-domain relationships by leveraging visual bounding box representation features to connect word features from two languages and to learn heterogeneous maps. MHCA and HCA facilitate cross-domain integration in the encoder through specialized heterogeneous attention mechanisms, enabling a single model to generate captions in two languages. We evaluate our approach on the MSCOCO dataset to generate captions in English and Chinese, two languages that exhibit significant differences in their language families. The experimental results demonstrate the superior performance of our method compared to existing advanced monolingual methods. Our proposed EHAT framework effectively addresses the challenges of cross-lingual image captioning, paving the way for improved multilingual image analysis and understanding.
Authors: Antonio Canela, Pol Caselles, Ibrar Malik, Eduard Ramon, Jaime Garc\'ia, Jordi S\'anchez-Riera, Gil Triginer, Francesc Moreno-Noguer
Abstract: Recent advances in full-head reconstruction have been obtained by optimizing a neural field through differentiable surface or volume rendering to represent a single scene. While these techniques achieve an unprecedented accuracy, they take several minutes, or even hours, due to the expensive optimization process required. In this work, we introduce InstantAvatar, a method that recovers full-head avatars from few images (down to just one) in a few seconds on commodity hardware. In order to speed up the reconstruction process, we propose a system that combines, for the first time, a voxel-grid neural field representation with a surface renderer. Notably, a naive combination of these two techniques leads to unstable optimizations that do not converge to valid solutions. In order to overcome this limitation, we present a novel statistical model that learns a prior distribution over 3D head signed distance functions using a voxel-grid based architecture. The use of this prior model, in combination with other design choices, results into a system that achieves 3D head reconstructions with comparable accuracy as the state-of-the-art with a 100x speed-up.
Authors: Liao Pan, Yang Feng, Wu Di, Liu Bo, Zhang Xingle
Abstract: In the field of multi-object tracking (MOT), recent Transformer based end-to-end models like MOTR have demonstrated exceptional performance on datasets such as DanceTracker. However, the computational demands of these models present challenges in training and deployment. Drawing inspiration from successful models like GPT, we present MO-YOLO, an efficient and computationally frugal end-to-end MOT model. MO-YOLO integrates principles from You Only Look Once (YOLO) and RT-DETR, adopting a decoder-only approach. By leveraging the decoder from RT-DETR and architectural components from YOLOv8, MO-YOLO achieves high speed, shorter training times, and proficient MOT performance. On the Dancetrack, MO-YOLO not only matches MOTR's performance but also surpasses it, achieving over twice the frames per second (MOTR 9.5 FPS, MO-YOLO 19.6 FPS). Furthermore, MO-YOLO demonstrates significantly reduced training times and lower hardware requirements compared to MOTR. This research introduces a promising paradigm for efficient end-to-end MOT, emphasizing enhanced performance and resource efficiency.
Authors: Qiying Yu, Quan Sun, Xiaosong Zhang, Yufeng Cui, Fan Zhang, Yue Cao, Xinlong Wang, Jingjing Liu
Abstract: Large multimodal models demonstrate remarkable generalist ability to perform diverse multimodal tasks in a zero-shot manner. Large-scale web-based image-text pairs contribute fundamentally to this success, but suffer from excessive noise. Recent studies use alternative captions synthesized by captioning models and have achieved notable benchmark performance. However, our experiments reveal significant Scalability Deficiency and World Knowledge Loss issues in models trained with synthetic captions, which have been largely obscured by their initial benchmark success. Upon closer examination, we identify the root cause as the overly-simplified language structure and lack of knowledge details in existing synthetic captions. To provide higher-quality and more scalable multimodal pretraining data, we propose CapsFusion, an advanced framework that leverages large language models to consolidate and refine information from both web-based image-text pairs and synthetic captions. Extensive experiments show that CapsFusion captions exhibit remarkable all-round superiority over existing captions in terms of model performance (e.g., 18.8 and 18.3 improvements in CIDEr score on COCO and NoCaps), sample efficiency (requiring 11-16 times less computation than baselines), world knowledge depth, and scalability. These effectiveness, efficiency and scalability advantages position CapsFusion as a promising candidate for future scaling of LMM training.
Authors: Yipeng Gao, Zeyu Wang, Wei-Shi Zheng, Cihang Xie, Yuyin Zhou
Abstract: Contrastive learning has emerged as a promising paradigm for 3D open-world understanding, i.e., aligning point cloud representation to image and text embedding space individually. In this paper, we introduce MixCon3D, a simple yet effective method aiming to sculpt holistic 3D representation in contrastive language-image-3D pre-training. In contrast to point cloud only, we develop the 3D object-level representation from complementary perspectives, e.g., multi-view rendered images with the point cloud. Then, MixCon3D performs language-3D contrastive learning, comprehensively depicting real-world 3D objects and bolstering text alignment. Additionally, we pioneer the first thorough investigation of various training recipes for the 3D contrastive learning paradigm, building a solid baseline with improved performance. Extensive experiments conducted on three representative benchmarks reveal that our method significantly improves over the baseline, surpassing the previous state-of-the-art performance on the challenging 1,156-category Objaverse-LVIS dataset by 5.7%. The versatility of MixCon3D is showcased in applications such as text-to-3D retrieval and point cloud captioning, further evidencing its efficacy in diverse scenarios. The code is available at https://github.com/UCSC-VLAA/MixCon3D.
Authors: Peng Jin, Ryuichi Takanobu, Wancai Zhang, Xiaochun Cao, Li Yuan
Abstract: Large language models have demonstrated impressive universal capabilities across a wide range of open-ended tasks and have extended their utility to encompass multimodal conversations. However, existing methods encounter challenges in effectively handling both image and video understanding, particularly with limited visual tokens. In this work, we introduce Chat-UniVi, a Unified Vision-language model capable of comprehending and engaging in conversations involving images and videos through a unified visual representation. Specifically, we employ a set of dynamic visual tokens to uniformly represent images and videos. This representation framework empowers the model to efficiently utilize a limited number of visual tokens to simultaneously capture the spatial details necessary for images and the comprehensive temporal relationship required for videos. Moreover, we leverage a multi-scale representation, enabling the model to perceive both high-level semantic concepts and low-level visual details. Notably, Chat-UniVi is trained on a mixed dataset containing both images and videos, allowing direct application to tasks involving both mediums without requiring any modifications. Extensive experimental results demonstrate that Chat-UniVi consistently outperforms even existing methods exclusively designed for either images or videos. Code is available at https://github.com/PKU-YuanGroup/Chat-UniVi.
Authors: Gonzalo J. Aniano Porcile, Jack Gindi, Shivansh Mundra, James R. Verbus, Hany Farid
Abstract: AI-based image generation has continued to rapidly improve, producing increasingly more realistic images with fewer obvious visual flaws. AI-generated images are being used to create fake online profiles which in turn are being used for spam, fraud, and disinformation campaigns. As the general problem of detecting any type of manipulated or synthesized content is receiving increasing attention, here we focus on a more narrow task of distinguishing a real face from an AI-generated face. This is particularly applicable when tackling inauthentic online accounts with a fake user profile photo. We show that by focusing on only faces, a more resilient and general-purpose artifact can be detected that allows for the detection of AI-generated faces from a variety of GAN- and diffusion-based synthesis engines, and across image resolutions (as low as 128 x 128 pixels) and qualities.
Authors: Ioannis Kakogeorgiou, Spyros Gidaris, Konstantinos Karantzalos, Nikos Komodakis
Abstract: Unsupervised object-centric learning aims to decompose scenes into interpretable object entities, termed slots. Slot-based auto-encoders stand out as a prominent method for this task. Within them, crucial aspects include guiding the encoder to generate object-specific slots and ensuring the decoder utilizes them during reconstruction. This work introduces two novel techniques, (i) an attention-based self-training approach, which distills superior slot-based attention masks from the decoder to the encoder, enhancing object segmentation, and (ii) an innovative patch-order permutation strategy for autoregressive transformers that strengthens the role of slot vectors in reconstruction. The effectiveness of these strategies is showcased experimentally. The combined approach significantly surpasses prior slot-based autoencoder methods in unsupervised object segmentation, especially with complex real-world images. We provide the implementation code at https://github.com/gkakogeorgiou/spot .
Authors: Jaime Corsetti, Davide Boscaini, Changjae Oh, Andrea Cavallaro, Fabio Poiesi
Abstract: We introduce the new setting of open-vocabulary object 6D pose estimation, in which a textual prompt is used to specify the object of interest. In contrast to existing approaches, in our setting (i) the object of interest is specified solely through the textual prompt, (ii) no object model (e.g., CAD or video sequence) is required at inference, and (iii) the object is imaged from two RGBD viewpoints of different scenes. To operate in this setting, we introduce a novel approach that leverages a Vision-Language Model to segment the object of interest from the scenes and to estimate its relative 6D pose. The key of our approach is a carefully devised strategy to fuse object-level information provided by the prompt with local image features, resulting in a feature space that can generalize to novel concepts. We validate our approach on a new benchmark based on two popular datasets, REAL275 and Toyota-Light, which collectively encompass 34 object instances appearing in four thousand image pairs. The results demonstrate that our approach outperforms both a well-established hand-crafted method and a recent deep learning-based baseline in estimating the relative 6D pose of objects in different scenes. Code and dataset are available at https://jcorsetti.github.io/oryon.
Authors: Vasileios Baltatzis, Rolandos Alexandros Potamias, Evangelos Ververas, Guanxiong Sun, Jiankang Deng, Stefanos Zafeiriou
Abstract: Sign Languages (SL) serve as the primary mode of communication for the Deaf and Hard of Hearing communities. Deep learning methods for SL recognition and translation have achieved promising results. However, Sign Language Production (SLP) poses a challenge as the generated motions must be realistic and have precise semantic meaning. Most SLP methods rely on 2D data, which hinders their realism. In this work, a diffusion-based SLP model is trained on a curated large-scale dataset of 4D signing avatars and their corresponding text transcripts. The proposed method can generate dynamic sequences of 3D avatars from an unconstrained domain of discourse using a diffusion process formed on a novel and anatomically informed graph neural network defined on the SMPL-X body skeleton. Through quantitative and qualitative experiments, we show that the proposed method considerably outperforms previous methods of SLP. This work makes an important step towards realistic neural sign avatars, bridging the communication gap between Deaf and hearing communities.
Authors: Yushi Hu, Otilia Stretcu, Chun-Ta Lu, Krishnamurthy Viswanathan, Kenji Hata, Enming Luo, Ranjay Krishna, Ariel Fuxman
Abstract: Solving complex visual tasks such as "Who invented the musical instrument on the right?" involves a composition of skills: understanding space, recognizing instruments, and also retrieving prior knowledge. Recent work shows promise by decomposing such tasks using a large language model (LLM) into an executable program that invokes specialized vision models. However, generated programs are error-prone: they omit necessary steps, include spurious ones, and are unable to recover when the specialized models give incorrect outputs. Moreover, they require loading multiple models, incurring high latency and computation costs. We propose Visual Program Distillation (VPD), an instruction tuning framework that produces a vision-language model (VLM) capable of solving complex visual tasks with a single forward pass. VPD distills the reasoning ability of LLMs by using them to sample multiple candidate programs, which are then executed and verified to identify a correct one. It translates each correct program into a language description of the reasoning steps, which are then distilled into a VLM. Extensive experiments show that VPD improves the VLM's ability to count, understand spatial relations, and reason compositionally. Our VPD-trained PaLI-X outperforms all prior VLMs, achieving state-of-the-art performance across complex vision tasks, including MMBench, OK-VQA, A-OKVQA, TallyQA, POPE, and Hateful Memes. An evaluation with human annotators also confirms that VPD improves model response factuality and consistency. Finally, experiments on content moderation demonstrate that VPD is also helpful for adaptation to real-world applications with limited data.
Authors: Adam Lilja, Junsheng Fu, Erik Stenborg, Lars Hammarstrand
Abstract: The task of online mapping is to predict a local map using current sensor observations, e.g. from lidar and camera, without relying on a pre-built map. State-of-the-art methods are based on supervised learning and are trained predominantly using two datasets: nuScenes and Argoverse 2. However, these datasets revisit the same geographic locations across training, validation, and test sets. Specifically, over $80$% of nuScenes and $40$% of Argoverse 2 validation and test samples are less than $5$ m from a training sample. At test time, the methods are thus evaluated more on how well they localize within a memorized implicit map built from the training data than on extrapolating to unseen locations. Naturally, this data leakage causes inflated performance numbers and we propose geographically disjoint data splits to reveal the true performance in unseen environments. Experimental results show that methods perform considerably worse, some dropping more than $45$ mAP, when trained and evaluated on proper data splits. Additionally, a reassessment of prior design choices reveals diverging conclusions from those based on the original split. Notably, the impact of lifting methods and the support from auxiliary tasks (e.g., depth supervision) on performance appears less substantial or follows a different trajectory than previously perceived. Splits can be found at https://github.com/LiljaAdam/geographical-splits
Authors: Nikita Starodubcev, Artem Fedorov, Artem Babenko, Dmitry Baranchuk
Abstract: Knowledge distillation methods have recently shown to be a promising direction to speedup the synthesis of large-scale diffusion models by requiring only a few inference steps. While several powerful distillation methods were recently proposed, the overall quality of student samples is typically lower compared to the teacher ones, which hinders their practical usage. In this work, we investigate the relative quality of samples produced by the teacher text-to-image diffusion model and its distilled student version. As our main empirical finding, we discover that a noticeable portion of student samples exhibit superior fidelity compared to the teacher ones, despite the "approximate" nature of the student. Based on this finding, we propose an adaptive collaboration between student and teacher diffusion models for effective text-to-image synthesis. Specifically, the distilled model produces the initial sample, and then an oracle decides whether it needs further improvements with a slow teacher model. Extensive experiments demonstrate that the designed pipeline surpasses state-of-the-art text-to-image alternatives for various inference budgets in terms of human preference. Furthermore, the proposed approach can be naturally used in popular applications such as text-guided image editing and controllable generation.
Authors: David Charatan, Sizhe Li, Andrea Tagliasacchi, Vincent Sitzmann
Abstract: We introduce pixelSplat, a feed-forward model that learns to reconstruct 3D radiance fields parameterized by 3D Gaussian primitives from pairs of images. Our model features real-time and memory-efficient rendering for scalable training as well as fast 3D reconstruction at inference time. To overcome local minima inherent to sparse and locally supported representations, we predict a dense probability distribution over 3D and sample Gaussian means from that probability distribution. We make this sampling operation differentiable via a reparameterization trick, allowing us to back-propagate gradients through the Gaussian splatting representation. We benchmark our method on wide-baseline novel view synthesis on the real-world RealEstate10k and ACID datasets, where we outperform state-of-the-art light field transformers and accelerate rendering by 2.5 orders of magnitude while reconstructing an interpretable and editable 3D radiance field.
Authors: David Pujol-Perich, Albert Clap\'es, Sergio Escalera
Abstract: Temporal Action Localization (TAL) is a complex task that poses relevant challenges, particularly when attempting to generalize on new -- unseen -- domains in real-world applications. These scenarios, despite realistic, are often neglected in the literature, exposing these solutions to important performance degradation. In this work, we tackle this issue by introducing, for the first time, an approach for Unsupervised Domain Adaptation (UDA) in sparse TAL, which we refer to as Semantic Adversarial unsupervised Domain Adaptation (SADA). Our contributions are threefold: (1) we pioneer the development of a domain adaptation model that operates on realistic sparse action detection benchmarks; (2) we tackle the limitations of global-distribution alignment techniques by introducing a novel adversarial loss that is sensitive to local class distributions, ensuring finer-grained adaptation; and (3) we present a novel set of benchmarks based on EpicKitchens100 and CharadesEgo, that evaluate multiple domain shifts in a comprehensive manner. Our experiments indicate that SADA improves the adaptation across domains when compared to fully supervised state-of-the-art and alternative UDA methods, attaining a performance boost of up to 6.14% mAP.
Authors: Tzofi Klinghoffer, Xiaoyu Xiang, Siddharth Somasundaram, Yuchen Fan, Christian Richardt, Ramesh Raskar, Rakesh Ranjan
Abstract: 3D reconstruction from a single-view is challenging because of the ambiguity from monocular cues and lack of information about occluded regions. Neural radiance fields (NeRF), while popular for view synthesis and 3D reconstruction, are typically reliant on multi-view images. Existing methods for single-view 3D reconstruction with NeRF rely on either data priors to hallucinate views of occluded regions, which may not be physically accurate, or shadows observed by RGB cameras, which are difficult to detect in ambient light and low albedo backgrounds. We propose using time-of-flight data captured by a single-photon avalanche diode to overcome these limitations. Our method models two-bounce optical paths with NeRF, using lidar transient data for supervision. By leveraging the advantages of both NeRF and two-bounce light measured by lidar, we demonstrate that we can reconstruct visible and occluded geometry without data priors or reliance on controlled ambient lighting or scene albedo. In addition, we demonstrate improved generalization under practical constraints on sensor spatial- and temporal-resolution. We believe our method is a promising direction as single-photon lidars become ubiquitous on consumer devices, such as phones, tablets, and headsets.
Authors: Zhan Li, Zhang Chen, Zhong Li, Yi Xu
Abstract: Novel view synthesis of dynamic scenes has been an intriguing yet challenging problem. Despite recent advancements, simultaneously achieving high-resolution photorealistic results, real-time rendering, and compact storage remains a formidable task. To address these challenges, we propose Spacetime Gaussian Feature Splatting as a novel dynamic scene representation, composed of three pivotal components. First, we formulate expressive Spacetime Gaussians by enhancing 3D Gaussians with temporal opacity and parametric motion/rotation. This enables Spacetime Gaussians to capture static, dynamic, as well as transient content within a scene. Second, we introduce splatted feature rendering, which replaces spherical harmonics with neural features. These features facilitate the modeling of view- and time-dependent appearance while maintaining small size. Third, we leverage the guidance of training error and coarse depth to sample new Gaussians in areas that are challenging to converge with existing pipelines. Experiments on several established real-world datasets demonstrate that our method achieves state-of-the-art rendering quality and speed, while retaining compact storage. At 8K resolution, our lite-version model can render at 60 FPS on an Nvidia RTX 4090 GPU. Our code is available at https://github.com/oppo-us-research/SpacetimeGaussians.
URLs: https://github.com/oppo-us-research/SpacetimeGaussians.
Authors: Zhengqing Yuan, Zhaoxu Li, Weiran Huang, Yanfang Ye, Lichao Sun
Abstract: In recent years, multimodal large language models (MLLMs) such as GPT-4V have demonstrated remarkable advancements, excelling in a variety of vision-language tasks. Despite their prowess, the closed-source nature and computational demands of such models limit their accessibility and applicability. This study introduces TinyGPT-V, a novel open-source MLLM, designed for efficient training and inference across various vision-language tasks, including image captioning (IC) and visual question answering (VQA). Leveraging a compact yet powerful architecture, TinyGPT-V integrates the Phi-2 language model with pre-trained vision encoders, utilizing a unique mapping module for visual and linguistic information fusion. With a training regimen optimized for small backbones and employing a diverse dataset amalgam, TinyGPT-V requires significantly lower computational resources 24GB for training and as little as 8GB for inference without compromising on performance. Our experiments demonstrate that TinyGPT-V, with its language model 2.8 billion parameters, achieves comparable results in VQA and image inference tasks to its larger counterparts while being uniquely suited for deployment on resource-constrained devices through innovative quantization techniques. This work not only paves the way for more accessible and efficient MLLMs but also underscores the potential of smaller, optimized models in bridging the gap between high performance and computational efficiency in real-world applications. Additionally, this paper introduces a new approach to multimodal large language models using smaller backbones. Our code and training weights are available in \url{https://github.com/DLYuanGod/TinyGPT-V}.
Authors: Zitong Huang, Ze Chen, Zhixing Chen, Erjin Zhou, Xinxing Xu, Rick Siow Mong Goh, Yong Liu, Wangmeng Zuo, Chunmei Feng
Abstract: Few-shot Class-Incremental Learning (FSCIL) aims to continuously learn new classes based on very limited training data without forgetting the old ones encountered. Existing studies solely relied on pure visual networks, while in this paper we solved FSCIL by leveraging the Vision-Language model (e.g., CLIP) and propose a simple yet effective framework, named Learning Prompt with Distribution-based Feature Replay (LP-DiF). We observe that simply using CLIP for zero-shot evaluation can substantially outperform the most influential methods. Then, prompt tuning technique is involved to further improve its adaptation ability, allowing the model to continually capture specific knowledge from each session. To prevent the learnable prompt from forgetting old knowledge in the new session, we propose a pseudo-feature replay approach. Specifically, we preserve the old knowledge of each class by maintaining a feature-level Gaussian distribution with a diagonal covariance matrix, which is estimated by the image features of training images and synthesized features generated from a VAE. When progressing to a new session, pseudo-features are sampled from old-class distributions combined with training images of the current session to optimize the prompt, thus enabling the model to learn new knowledge while retaining old knowledge. Experiments on three prevalent benchmarks, i.e., CIFAR100, mini-ImageNet, CUB-200, and two more challenging benchmarks, i.e., SUN-397 and CUB-200$^*$ proposed in this paper showcase the superiority of LP-DiF, achieving new state-of-the-art (SOTA) in FSCIL. Code is publicly available at https://github.com/1170300714/LP-DiF.
Authors: Tim Michels, Daniel M\"ackelmann, Reinhard Koch
Abstract: Among the common applications of plenoptic cameras are depth reconstruction and post-shot refocusing. These require a calibration relating the camera-side light field to that of the scene. Numerous methods with this goal have been developed based on thin lens models for the plenoptic camera's main lens and microlenses. Our work addresses the often-overlooked role of the main lens exit pupil in these models and specifically in the decoding process of standard plenoptic camera (SPC) images. We formally deduce the connection between the refocusing distance and the resampling parameter for the decoded light field and provide an analysis of the errors that arise when the exit pupil is not considered. In addition, previous work is revisited with respect to the exit pupil's role and all theoretical results are validated through a ray-tracing-based simulation. With the public release of the evaluated SPC designs alongside our simulation and experimental data we aim to contribute to a more accurate and nuanced understanding of plenoptic camera optics.
Authors: Nikola Zubi\'c, Mathias Gehrig, Davide Scaramuzza
Abstract: Today, state-of-the-art deep neural networks that process event-camera data first convert a temporal window of events into dense, grid-like input representations. As such, they exhibit poor generalizability when deployed at higher inference frequencies (i.e., smaller temporal windows) than the ones they were trained on. We address this challenge by introducing state-space models (SSMs) with learnable timescale parameters to event-based vision. This design adapts to varying frequencies without the need to retrain the network at different frequencies. Additionally, we investigate two strategies to counteract aliasing effects when deploying the model at higher frequencies. We comprehensively evaluate our approach against existing methods based on RNN and Transformer architectures across various benchmarks, including Gen1 and 1 Mpx event camera datasets. Our results demonstrate that SSM-based models train 33% faster and also exhibit minimal performance degradation when tested at higher frequencies than the training input. Traditional RNN and Transformer models exhibit performance drops of more than 20 mAP, with SSMs having a drop of 3.31 mAP, highlighting the effectiveness of SSMs in event-based vision tasks.
Authors: Hao Li, Ying Chen, Yifei Chen, Wenxian Yang, Bowen Ding, Yuchen Han, Liansheng Wang, Rongshan Yu
Abstract: Whole Slide Image (WSI) classification is often formulated as a Multiple Instance Learning (MIL) problem. Recently, Vision-Language Models (VLMs) have demonstrated remarkable performance in WSI classification. However, existing methods leverage coarse-grained pathogenetic descriptions for visual representation supervision, which are insufficient to capture the complex visual appearance of pathogenetic images, hindering the generalizability of models on diverse downstream tasks. Additionally, processing high-resolution WSIs can be computationally expensive. In this paper, we propose a novel "Fine-grained Visual-Semantic Interaction" (FiVE) framework for WSI classification. It is designed to enhance the model's generalizability by leveraging the interaction between localized visual patterns and fine-grained pathological semantics. Specifically, with meticulously designed queries, we start by utilizing a large language model to extract fine-grained pathological descriptions from various non-standardized raw reports. The output descriptions are then reconstructed into fine-grained labels used for training. By introducing a Task-specific Fine-grained Semantics (TFS) module, we enable prompts to capture crucial visual information in WSIs, which enhances representation learning and augments generalization capabilities significantly. Furthermore, given that pathological visual patterns are redundantly distributed across tissue slices, we sample a subset of visual instances during training. Our method demonstrates robust generalizability and strong transferability, dominantly outperforming the counterparts on the TCGA Lung Cancer dataset with at least 9.19% higher accuracy in few-shot experiments. The code is available at: https://github.com/ls1rius/WSI_FiVE.
Authors: Alexander C. Jenke, Sebastian Bodenstedt, Fiona R. Kolbinger, Marius Distler, J\"urgen Weitz, Stefanie Speidel
Abstract: Understanding a surgical scene is crucial for computer-assisted surgery systems to provide any intelligent assistance functionality. One way of achieving this scene understanding is via scene segmentation, where every pixel of a frame is classified and therefore identifies the visible structures and tissues. Progress on fully segmenting surgical scenes has been made using machine learning. However, such models require large amounts of annotated training data, containing examples of all relevant object classes. Such fully annotated datasets are hard to create, as every pixel in a frame needs to be annotated by medical experts and, therefore, are rarely available. In this work, we propose a method to combine multiple partially annotated datasets, which provide complementary annotations, into one model, enabling better scene segmentation and the use of multiple readily available datasets. Our method aims to combine available data with complementary labels by leveraging mutual exclusive properties to maximize information. Specifically, we propose to use positive annotations of other classes as negative samples and to exclude background pixels of binary annotations, as we cannot tell if they contain a class not annotated but predicted by the model. We evaluate our method by training a DeepLabV3 on the publicly available Dresden Surgical Anatomy Dataset, which provides multiple subsets of binary segmented anatomical structures. Our approach successfully combines 6 classes into one model, increasing the overall Dice Score by 4.4% compared to an ensemble of models trained on the classes individually. By including information on multiple classes, we were able to reduce confusion between stomach and colon by 24%. Our results demonstrate the feasibility of training a model on multiple datasets. This paves the way for future work further alleviating the need for one large, fully segmented datasets.
Authors: Taeheon Kim, Sebin Shin, Youngjoon Yu, Hak Gu Kim, Yong Man Ro
Abstract: RGBT multispectral pedestrian detection has emerged as a promising solution for safety-critical applications that require day/night operations. However, the modality bias problem remains unsolved as multispectral pedestrian detectors learn the statistical bias in datasets. Specifically, datasets in multispectral pedestrian detection mainly distribute between ROTO (day) and RXTO (night) data; the majority of the pedestrian labels statistically co-occur with their thermal features. As a result, multispectral pedestrian detectors show poor generalization ability on examples beyond this statistical correlation, such as ROTX data. To address this problem, we propose a novel Causal Mode Multiplexer (CMM) framework that effectively learns the causalities between multispectral inputs and predictions. Moreover, we construct a new dataset (ROTX-MP) to evaluate modality bias in multispectral pedestrian detection. ROTX-MP mainly includes ROTX examples not presented in previous datasets. Extensive experiments demonstrate that our proposed CMM framework generalizes well on existing datasets (KAIST, CVC-14, FLIR) and the new ROTX-MP. We will release our new dataset to the public for future research.
Authors: Xin Zhou, Dingkang Liang, Wei Xu, Xingkui Zhu, Yihan Xu, Zhikang Zou, Xiang Bai
Abstract: Point cloud analysis has achieved outstanding performance by transferring point cloud pre-trained models. However, existing methods for model adaptation usually update all model parameters, i.e., full fine-tuning paradigm, which is inefficient as it relies on high computational costs (e.g., training GPU memory) and massive storage space. In this paper, we aim to study parameter-efficient transfer learning for point cloud analysis with an ideal trade-off between task performance and parameter efficiency. To achieve this goal, we freeze the parameters of the default pre-trained models and then propose the Dynamic Adapter, which generates a dynamic scale for each token, considering the token significance to the downstream task. We further seamlessly integrate Dynamic Adapter with Prompt Tuning (DAPT) by constructing Internal Prompts, capturing the instance-specific features for interaction. Extensive experiments conducted on five challenging datasets demonstrate that the proposed DAPT achieves superior performance compared to the full fine-tuning counterparts while significantly reducing the trainable parameters and training GPU memory by 95% and 35%, respectively. Code is available at https://github.com/LMD0311/DAPT.
Authors: Chanyoung Kim, Woojung Han, Dayun Ju, Seong Jae Hwang
Abstract: Semantic segmentation has innately relied on extensive pixel-level annotated data, leading to the emergence of unsupervised methodologies. Among them, leveraging self-supervised Vision Transformers for unsupervised semantic segmentation (USS) has been making steady progress with expressive deep features. Yet, for semantically segmenting images with complex objects, a predominant challenge remains: the lack of explicit object-level semantic encoding in patch-level features. This technical limitation often leads to inadequate segmentation of complex objects with diverse structures. To address this gap, we present a novel approach, EAGLE, which emphasizes object-centric representation learning for unsupervised semantic segmentation. Specifically, we introduce EiCue, a spectral technique providing semantic and structural cues through an eigenbasis derived from the semantic similarity matrix of deep image features and color affinity from an image. Further, by incorporating our object-centric contrastive loss with EiCue, we guide our model to learn object-level representations with intra- and inter-image object-feature consistency, thereby enhancing semantic accuracy. Extensive experiments on COCO-Stuff, Cityscapes, and Potsdam-3 datasets demonstrate the state-of-the-art USS results of EAGLE with accurate and consistent semantic segmentation across complex scenes.
Authors: Geonho Bang, Kwangjin Choi, Jisong Kim, Dongsuk Kum, Jun Won Choi
Abstract: The inherent noisy and sparse characteristics of radar data pose challenges in finding effective representations for 3D object detection. In this paper, we propose RadarDistill, a novel knowledge distillation (KD) method, which can improve the representation of radar data by leveraging LiDAR data. RadarDistill successfully transfers desirable characteristics of LiDAR features into radar features using three key components: Cross-Modality Alignment (CMA), Activation-based Feature Distillation (AFD), and Proposal-based Feature Distillation (PFD). CMA enhances the density of radar features by employing multiple layers of dilation operations, effectively addressing the challenge of inefficient knowledge transfer from LiDAR to radar. AFD selectively transfers knowledge based on regions of the LiDAR features, with a specific focus on areas where activation intensity exceeds a predefined threshold. PFD similarly guides the radar network to selectively mimic features from the LiDAR network within the object proposals. Our comparative analyses conducted on the nuScenes datasets demonstrate that RadarDistill achieves state-of-the-art (SOTA) performance for radar-only object detection task, recording 20.5% in mAP and 43.7% in NDS. Also, RadarDistill significantly improves the performance of the camera-radar fusion model.
Authors: Baran Ozaydin, Tong Zhang, Deblina Bhattacharjee, Sabine S\"usstrunk, Mathieu Salzmann
Abstract: Unsupervised Semantic Segmentation (USS) involves segmenting images without relying on predefined labels, aiming to alleviate the burden of extensive human labeling. Existing methods utilize features generated by self-supervised models and specific priors for clustering. However, their clustering objectives are not involved in the optimization of the features during training. Additionally, due to the lack of clear class definitions in USS, the resulting segments may not align well with the clustering objective. In this paper, we introduce a novel approach called Optimally Matched Hierarchy (OMH) to simultaneously address the above issues. The core of our method lies in imposing structured sparsity on the feature space, which allows the features to encode information with different levels of granularity. The structure of this sparsity stems from our hierarchy (OMH). To achieve this, we learn a soft but sparse hierarchy among parallel clusters through Optimal Transport. Our OMH yields better unsupervised segmentation performance compared to existing USS methods. Our extensive experiments demonstrate the benefits of OMH when utilizing our differentiable paradigm. We will make our code publicly available.
Authors: Zhuoxuan Peng, S. -H. Gary Chan
Abstract: Due to its promising results, density map regression has been widely employed for image-based crowd counting. The approach, however, often suffers from severe performance degradation when tested on data from unseen scenarios, the so-called "domain shift" problem. To address the problem, we investigate in this work single domain generalization (SDG) for crowd counting. The existing SDG approaches are mainly for image classification and segmentation, and can hardly be extended to our case due to its regression nature and label ambiguity (i.e., ambiguous pixel-level ground truths). We propose MPCount, a novel effective SDG approach even for narrow source distribution. MPCount stores diverse density values for density map regression and reconstructs domain-invariant features by means of only one memory bank, a content error mask and attention consistency loss. By partitioning the image into grids, it employs patch-wise classification as an auxiliary task to mitigate label ambiguity. Through extensive experiments on different datasets, MPCount is shown to significantly improve counting accuracy compared to the state of the art under diverse scenarios unobserved in the training data characterized by narrow source distribution. Code is available at https://github.com/Shimmer93/MPCount.
Authors: Hala Djeghim, Nathan Piasco, Moussab Bennehar, Luis Rold\~ao, Dzmitry Tsishkou, D\'esir\'e Sidib\'e
Abstract: Neural implicit surface representation methods have recently shown impressive 3D reconstruction results. However, existing solutions struggle to reconstruct urban outdoor scenes due to their large, unbounded, and highly detailed nature. Hence, to achieve accurate reconstructions, additional supervision data such as LiDAR, strong geometric priors, and long training times are required. To tackle such issues, we present SCILLA, a new hybrid implicit surface learning method to reconstruct large driving scenes from 2D images. SCILLA's hybrid architecture models two separate implicit fields: one for the volumetric density and another for the signed distance to the surface. To accurately represent urban outdoor scenarios, we introduce a novel volume-rendering strategy that relies on self-supervised probabilistic density estimation to sample points near the surface and transition progressively from volumetric to surface representation. Our solution permits a proper and fast initialization of the signed distance field without relying on any geometric prior on the scene, compared to concurrent methods. By conducting extensive experiments on four outdoor driving datasets, we show that SCILLA can learn an accurate and detailed 3D surface scene representation in various urban scenarios while being two times faster to train compared to previous state-of-the-art solutions.
Authors: Hiba Dahmani, Moussab Bennehar, Nathan Piasco, Luis Roldao, Dzmitry Tsishkou
Abstract: Implicit neural representation methods have shown impressive advancements in learning 3D scenes from unstructured in-the-wild photo collections but are still limited by the large computational cost of volumetric rendering. More recently, 3D Gaussian Splatting emerged as a much faster alternative with superior rendering quality and training efficiency, especially for small-scale and object-centric scenarios. Nevertheless, this technique suffers from poor performance on unstructured in-the-wild data. To tackle this, we extend over 3D Gaussian Splatting to handle unstructured image collections. We achieve this by modeling appearance to seize photometric variations in the rendered images. Additionally, we introduce a new mechanism to train transient Gaussians to handle the presence of scene occluders in an unsupervised manner. Experiments on diverse photo collection scenes and multi-pass acquisition of outdoor landmarks show the effectiveness of our method over prior works achieving state-of-the-art results with improved efficiency.
Authors: Runwei Guan, Liye Jia, Fengyufan Yang, Shanliang Yao, Erick Purwanto, Xiaohui Zhu, Eng Gee Lim, Jeremy Smith, Ka Lok Man, Xuming Hu, Yutao Yue
Abstract: The perception of waterways based on human intent is significant for autonomous navigation and operations of Unmanned Surface Vehicles (USVs) in water environments. Inspired by visual grounding, we introduce WaterVG, the first visual grounding dataset designed for USV-based waterway perception based on human prompts. WaterVG encompasses prompts describing multiple targets, with annotations at the instance level including bounding boxes and masks. Notably, WaterVG includes 11,568 samples with 34,987 referred targets, whose prompts integrates both visual and radar characteristics. The pattern of text-guided two sensors equips a finer granularity of text prompts with visual and radar features of referred targets. Moreover, we propose a low-power visual grounding model, Potamoi, which is a multi-task model with a well-designed Phased Heterogeneous Modality Fusion (PHMF) mode, including Adaptive Radar Weighting (ARW) and Multi-Head Slim Cross Attention (MHSCA). Exactly, ARW extracts required radar features to fuse with vision for prompt alignment. MHSCA is an efficient fusion module with a remarkably small parameter count and FLOPs, elegantly fusing scenario context captured by two sensors with linguistic features, which performs expressively on visual grounding tasks. Comprehensive experiments and evaluations have been conducted on WaterVG, where our Potamoi archives state-of-the-art performances compared with counterparts.
Authors: Bowen Zhang, Yiji Cheng, Jiaolong Yang, Chunyu Wang, Feng Zhao, Yansong Tang, Dong Chen, Baining Guo
Abstract: 3D Gaussian Splatting (GS) have achieved considerable improvement over Neural Radiance Fields in terms of 3D fitting fidelity and rendering speed. However, this unstructured representation with scattered Gaussians poses a significant challenge for generative modeling. To address the problem, we introduce GaussianCube, a structured GS representation that is both powerful and efficient for generative modeling. We achieve this by first proposing a modified densification-constrained GS fitting algorithm which can yield high-quality fitting results using a fixed number of free Gaussians, and then re-arranging the Gaussians into a predefined voxel grid via Optimal Transport. The structured grid representation allows us to use standard 3D U-Net as our backbone in diffusion generative modeling without elaborate designs. Extensive experiments conducted on ShapeNet and OmniObject3D show that our model achieves state-of-the-art generation results both qualitatively and quantitatively, underscoring the potential of GaussianCube as a powerful and versatile 3D representation.
Authors: Hu Gao, Depeng Dang
Abstract: Image deblurring aims to restore a high-quality image from its corresponding blurred. The emergence of CNNs and Transformers has enabled significant progress. However, these methods often face the dilemma between eliminating long-range degradation perturbations and maintaining computational efficiency. While the selective state space model (SSM) shows promise in modeling long-range dependencies with linear complexity, it also encounters challenges such as local pixel forgetting and channel redundancy. To address this issue, we propose an efficient image deblurring network that leverages selective state spaces model to aggregate enriched and accurate features. Specifically, we introduce an aggregate local and global information block (ALGBlock) designed to effectively capture and integrate both local invariant properties and non-local information. The ALGBlock comprises two primary modules: a module for capturing local and global features (CLGF), and a feature aggregation module (FA). The CLGF module is composed of two branches: the global branch captures long-range dependency features via a selective state spaces model, while the local branch employs simplified channel attention to model local connectivity, thereby reducing local pixel forgetting and channel redundancy. In addition, we design a FA module to accentuate the local part by recalibrating the weight during the aggregation of the two branches for restoration. Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches on widely used benchmarks.
Authors: Yikang Zhou, Tao Zhang, Shunping Ji, Shuicheng Yan, Xiangtai Li
Abstract: Modern video segmentation methods adopt object queries to perform inter-frame association and demonstrate satisfactory performance in tracking continuously appearing objects despite large-scale motion and transient occlusion. However, they all underperform on newly emerging and disappearing objects that are common in the real world because they attempt to model object emergence and disappearance through feature transitions between background and foreground queries that have significant feature gaps. We introduce Dynamic Anchor Queries (DAQ) to shorten the transition gap between the anchor and target queries by dynamically generating anchor queries based on the features of potential candidates. Furthermore, we introduce a query-level object Emergence and Disappearance Simulation (EDS) strategy, which unleashes DAQ's potential without any additional cost. Finally, we combine our proposed DAQ and EDS with DVIS to obtain DVIS-DAQ. Extensive experiments demonstrate that DVIS-DAQ achieves a new state-of-the-art (SOTA) performance on five mainstream video segmentation benchmarks. Code and models are available at \url{https://github.com/SkyworkAI/DAQ-VS}.
Authors: Amitangshu Mukherjee, Timur Ibrayev, Kaushik Roy
Abstract: Current Deep Neural Networks are vulnerable to adversarial examples, which alter their predictions by adding carefully crafted noise. Since human eyes are robust to such inputs, it is possible that the vulnerability stems from the standard way of processing inputs in one shot by processing every pixel with the same importance. In contrast, neuroscience suggests that the human vision system can differentiate salient features by (1) switching between multiple fixation points (saccades) and (2) processing the surrounding with a non-uniform external resolution (foveation). In this work, we advocate that the integration of such active vision mechanisms into current deep learning systems can offer robustness benefits. Specifically, we empirically demonstrate the inherent robustness of two active vision methods - GFNet and FALcon - under a black box threat model. By learning and inferencing based on downsampled glimpses obtained from multiple distinct fixation points within an input, we show that these active methods achieve (2-3) times greater robustness compared to a standard passive convolutional network under state-of-the-art adversarial attacks. More importantly, we provide illustrative and interpretable visualization analysis that demonstrates how performing inference from distinct fixation points makes active vision methods less vulnerable to malicious inputs.
Authors: Pietro Bonazzi
Abstract: Existing deep learning methods for the reconstruction and denoising of point clouds rely on small datasets of 3D shapes. We circumvent the problem by leveraging deep learning methods trained on billions of images. We propose a method to reconstruct point clouds from few images and to denoise point clouds from their rendering by exploiting prior knowledge distilled from image-based deep learning models. To improve reconstruction in constraint settings, we regularize the training of a differentiable renderer with hybrid surface and appearance by introducing semantic consistency supervision. In addition, we propose a pipeline to finetune Stable Diffusion to denoise renderings of noisy point clouds and we demonstrate how these learned filters can be used to remove point cloud noise coming without 3D supervision. We compare our method with DSS and PointRadiance and achieved higher quality 3D reconstruction on the Sketchfab Testset and SCUT Dataset.
Authors: Tao Hu, Fangzhou Hong, Zhaoxi Chen, Ziwei Liu
Abstract: We present FashionEngine, an interactive 3D human generation and editing system that allows us to design 3D digital humans in a way that aligns with how humans interact with the world, such as natural languages, visual perceptions, and hand-drawing. FashionEngine automates the 3D human production with three key components: 1) A pre-trained 3D human diffusion model that learns to model 3D humans in a semantic UV latent space from 2D image training data, which provides strong priors for diverse generation and editing tasks. 2) Multimodality-UV Space encoding the texture appearance, shape topology, and textual semantics of human clothing in a canonical UV-aligned space, which faithfully aligns the user multimodal inputs with the implicit UV latent space for controllable 3D human editing. The multimodality-UV space is shared across different user inputs, such as texts, images, and sketches, which enables various joint multimodal editing tasks. 3) Multimodality-UV Aligned Sampler learns to sample high-quality and diverse 3D humans from the diffusion prior for multimodal user inputs. Extensive experiments validate FashionEngine's state-of-the-art performance for conditional generation/editing tasks. In addition, we present an interactive user interface for our FashionEngine that enables both conditional and unconditional generation tasks, and editing tasks including pose/view/shape control, text-, image-, and sketch-driven 3D human editing and 3D virtual try-on, in a unified framework. Our project page is at: https://taohuumd.github.io/projects/FashionEngine.
Authors: Pietro Bonazzi
Abstract: Synthesizing realistic and diverse indoor 3D scene layouts in a controllable fashion opens up applications in simulated navigation and virtual reality. As concise and robust representations of a scene, scene graphs have proven to be well-suited as the semantic control on the generated layout. We present a variant of the conditional variational autoencoder (cVAE) model to synthesize 3D scenes from scene graphs and floor plans. We exploit the properties of self-attention layers to capture high-level relationships between objects in a scene, and use these as the building blocks of our model. Our model, leverages graph transformers to estimate the size, dimension and orientation of the objects in a room while satisfying relationships in the given scene graph. Our experiments shows self-attention layers leads to sparser (7.9x compared to Graphto3D) and more diverse scenes (16%).
Authors: Jinbae Im, JeongYeon Nam, Nokyung Park, Hyungmin Lee, Seunghyun Park
Abstract: Scene Graph Generation (SGG) is a challenging task of detecting objects and predicting relationships between objects. After DETR was developed, one-stage SGG models based on a one-stage object detector have been actively studied. However, complex modeling is used to predict the relationship between objects, and the inherent relationship between object queries learned in the multi-head self-attention of the object detector has been neglected. We propose a lightweight one-stage SGG model that extracts the relation graph from the various relationships learned in the multi-head self-attention layers of the DETR decoder. By fully utilizing the self-attention by-products, the relation graph can be extracted effectively with a shallow relation extraction head. Considering the dependency of the relation extraction task on the object detection task, we propose a novel relation smoothing technique that adjusts the relation label adaptively according to the quality of the detected objects. By the relation smoothing, the model is trained according to the continuous curriculum that focuses on object detection task at the beginning of training and performs multi-task learning as the object detection performance gradually improves. Furthermore, we propose a connectivity prediction task that predicts whether a relation exists between object pairs as an auxiliary task of the relation extraction. We demonstrate the effectiveness and efficiency of our method for the Visual Genome and Open Image V6 datasets. Our code is publicly available at https://github.com/naver-ai/egtr.
Authors: Fangzhou Mu, Sicheng Mo, Yin Li
Abstract: Temporal grounding of text descriptions in videos is a central problem in vision-language learning and video understanding. Existing methods often prioritize accuracy over scalability -- they have been optimized for grounding only a few text queries within short videos, and fail to scale up to long videos with hundreds of queries. In this paper, we study the effect of cross-modal fusion on the scalability of video grounding models. Our analysis establishes late fusion as a more cost-effective fusion scheme for long-form videos with many text queries. Moreover, it leads us to a novel, video-centric sampling scheme for efficient training. Based on these findings, we present SnAG, a simple baseline for scalable and accurate video grounding. Without bells and whistles, SnAG is 43% more accurate and 1.5x faster than CONE, a state of the art for long-form video grounding on the challenging MAD dataset, while achieving highly competitive results on short videos.
Authors: Haofan Wang, Matteo Spinelli, Qixun Wang, Xu Bai, Zekui Qin, Anthony Chen
Abstract: Tuning-free diffusion-based models have demonstrated significant potential in the realm of image personalization and customization. However, despite this notable progress, current models continue to grapple with several complex challenges in producing style-consistent image generation. Firstly, the concept of style is inherently underdetermined, encompassing a multitude of elements such as color, material, atmosphere, design, and structure, among others. Secondly, inversion-based methods are prone to style degradation, often resulting in the loss of fine-grained details. Lastly, adapter-based approaches frequently require meticulous weight tuning for each reference image to achieve a balance between style intensity and text controllability. In this paper, we commence by examining several compelling yet frequently overlooked observations. We then proceed to introduce InstantStyle, a framework designed to address these issues through the implementation of two key strategies: 1) A straightforward mechanism that decouples style and content from reference images within the feature space, predicated on the assumption that features within the same space can be either added to or subtracted from one another. 2) The injection of reference image features exclusively into style-specific blocks, thereby preventing style leaks and eschewing the need for cumbersome weight tuning, which often characterizes more parameter-heavy designs.Our work demonstrates superior visual stylization outcomes, striking an optimal balance between the intensity of style and the controllability of textual elements. Our codes will be available at https://github.com/InstantStyle/InstantStyle.
Authors: Zhihao Chen, Yiyuan Ge
Abstract: The goal of occluded person re-identification (ReID) is to retrieve specific pedestrians in occluded situations. However, occluded person ReID still suffers from background clutter and low-quality local feature representations, which limits model performance. In our research, we introduce a new framework called PAB-ReID, which is a novel ReID model incorporating part-attention mechanisms to tackle the aforementioned issues effectively. Firstly, we introduce the human parsing label to guide the generation of more accurate human part attention maps. In addition, we propose a fine-grained feature focuser for generating fine-grained human local feature representations while suppressing background interference. Moreover, We also design a part triplet loss to supervise the learning of human local features, which optimizes intra/inter-class distance. We conducted extensive experiments on specialized occlusion and regular ReID datasets, showcasing that our approach outperforms the existing state-of-the-art methods.
Authors: Ziyao Zeng, Daniel Wang, Fengyu Yang, Hyoungseob Park, Yangchao Wu, Stefano Soatto, Byung-Woo Hong, Dong Lao, Alex Wong
Abstract: Three-dimensional (3D) reconstruction from a single image is an ill-posed problem with inherent ambiguities, i.e. scale. Predicting a 3D scene from text description(s) is similarly ill-posed, i.e. spatial arrangements of objects described. We investigate the question of whether two inherently ambiguous modalities can be used in conjunction to produce metric-scaled reconstructions. To test this, we focus on monocular depth estimation, the problem of predicting a dense depth map from a single image, but with an additional text caption describing the scene. To this end, we begin by encoding the text caption as a mean and standard deviation; using a variational framework, we learn the distribution of the plausible metric reconstructions of 3D scenes corresponding to the text captions as a prior. To "select" a specific reconstruction or depth map, we encode the given image through a conditional sampler that samples from the latent space of the variational text encoder, which is then decoded to the output depth map. Our approach is trained alternatingly between the text and image branches: in one optimization step, we predict the mean and standard deviation from the text description and sample from a standard Gaussian, and in the other, we sample using a (image) conditional sampler. Once trained, we directly predict depth from the encoded text using the conditional sampler. We demonstrate our approach on indoor (NYUv2) and outdoor (KITTI) scenarios, where we show that language can consistently improve performance in both.
Authors: Nenad Marku\v{s}, Mirko Su\v{z}njevi\'c
Abstract: Recently there has been renewed interest in signed distance bound representations due to their unique properties for 3D shape modelling. This is especially the case for deep learning-based bounds. However, it is beneficial to work with polygons in most computer-graphics applications. Thus, in this paper we introduce and investigate an asymptotically fast method for transforming signed distance bounds into polygon meshes. This is achieved by combining the principles of sphere tracing (or ray marching) with traditional polygonization techniques, such as Marching Cubes. We provide theoretical and experimental evidence that this approach is of the $O(N^2\log N)$ computational complexity for a polygonization grid with $N^3$ cells. The algorithm is tested on both a set of primitive shapes as well as signed distance bounds generated from point clouds by machine learning (and represented as neural networks). Given its speed, implementation simplicity and portability, we argue that it could prove useful during the modelling stage as well as in shape compression for storage. The code is available here: https://github.com/nenadmarkus/gridhopping
Authors: Srijay Deshpande, Fayyaz Minhas, Nasir Rajpoot
Abstract: Generating realistic tissue images with annotations is a challenging task that is important in many computational histopathology applications. Synthetically generated images and annotations are valuable for training and evaluating algorithms in this domain. To address this, we propose an interactive framework generating pairs of realistic colorectal cancer histology images with corresponding glandular masks from glandular structure layouts. The framework accurately captures vital features like stroma, goblet cells, and glandular lumen. Users can control gland appearance by adjusting parameters such as the number of glands, their locations, and sizes. The generated images exhibit good Frechet Inception Distance (FID) scores compared to the state-of-the-art image-to-image translation model. Additionally, we demonstrate the utility of our synthetic annotations for evaluating gland segmentation algorithms. Furthermore, we present a methodology for constructing glandular masks using advanced deep generative models, such as latent diffusion models. These masks enable tissue image generation through a residual encoder-decoder network.
Authors: Zhe Huang, Benjamin S. Wessler, Michael C. Hughes
Abstract: Aortic stenosis (AS) is a degenerative valve condition that causes substantial morbidity and mortality. This condition is under-diagnosed and under-treated. In clinical practice, AS is diagnosed with expert review of transthoracic echocardiography, which produces dozens of ultrasound images of the heart. Only some of these views show the aortic valve. To automate screening for AS, deep networks must learn to mimic a human expert's ability to identify views of the aortic valve then aggregate across these relevant images to produce a study-level diagnosis. We find previous approaches to AS detection yield insufficient accuracy due to relying on inflexible averages across images. We further find that off-the-shelf attention-based multiple instance learning (MIL) performs poorly. We contribute a new end-to-end MIL approach with two key methodological innovations. First, a supervised attention technique guides the learned attention mechanism to favor relevant views. Second, a novel self-supervised pretraining strategy applies contrastive learning on the representation of the whole study instead of individual images as commonly done in prior literature. Experiments on an open-access dataset and an external validation set show that our approach yields higher accuracy while reducing model size.
Authors: Sergio Calvo-Ordonez, Chun-Wun Cheng, Jiahao Huang, Lipei Zhang, Guang Yang, Carola-Bibiane Schonlieb, Angelica I Aviles-Rivero
Abstract: Diffusion Probabilistic Models stand as a critical tool in generative modelling, enabling the generation of complex data distributions. This family of generative models yields record-breaking performance in tasks such as image synthesis, video generation, and molecule design. Despite their capabilities, their efficiency, especially in the reverse process, remains a challenge due to slow convergence rates and high computational costs. In this paper, we introduce an approach that leverages continuous dynamical systems to design a novel denoising network for diffusion models that is more parameter-efficient, exhibits faster convergence, and demonstrates increased noise robustness. Experimenting with Denoising Diffusion Probabilistic Models (DDPMs), our framework operates with approximately a quarter of the parameters, and $\sim$ 30\% of the Floating Point Operations (FLOPs) compared to standard U-Nets in DDPMs. Furthermore, our model is notably faster in inference than the baseline when measured in fair and equal conditions. We also provide a mathematical intuition as to why our proposed reverse process is faster as well as a mathematical discussion of the empirical tradeoffs in the denoising downstream task. Finally, we argue that our method is compatible with existing performance enhancement techniques, enabling further improvements in efficiency, quality, and speed.
Authors: Yao Sun, Yi Wang, Michael Eineder
Abstract: Quick and automated earthquake-damaged building detection from post-event satellite imagery is crucial, yet it is challenging due to the scarcity of training data required to develop robust algorithms. This letter presents the first dataset dedicated to detecting earthquake-damaged buildings from post-event very high resolution (VHR) Synthetic Aperture Radar (SAR) and optical imagery. Utilizing open satellite imagery and annotations acquired after the 2023 Turkey-Syria earthquakes, we deliver a dataset of coregistered building footprints and satellite image patches of both SAR and optical data, encompassing more than four thousand buildings. The task of damaged building detection is formulated as a binary image classification problem, that can also be treated as an anomaly detection problem due to extreme class imbalance. We provide baseline methods and results to serve as references for comparison. Researchers can utilize this dataset to expedite algorithm development, facilitating the rapid detection of damaged buildings in response to future events. The dataset and codes together with detailed explanations and visualization are made publicly available at \url{https://github.com/ya0-sun/PostEQ-SARopt-BuildingDamage}.
URLs: https://github.com/ya0-sun/PostEQ-SARopt-BuildingDamage
Authors: Eugenio Chisari, Nick Heppert, Tim Welschehold, Wolfram Burgard, Abhinav Valada
Abstract: Reliable object grasping is a crucial capability for autonomous robots. However, many existing grasping approaches focus on general clutter removal without explicitly modeling objects and thus only relying on the visible local geometry. We introduce CenterGrasp, a novel framework that combines object awareness and holistic grasping. CenterGrasp learns a general object prior by encoding shapes and valid grasps in a continuous latent space. It consists of an RGB-D image encoder that leverages recent advances to detect objects and infer their pose and latent code, and a decoder to predict shape and grasps for each object in the scene. We perform extensive experiments on simulated as well as real-world cluttered scenes and demonstrate strong scene reconstruction and 6-DoF grasp-pose estimation performance. Compared to the state of the art, CenterGrasp achieves an improvement of 38.5 mm in shape reconstruction and 33 percentage points on average in grasp success. We make the code and trained models publicly available at http://centergrasp.cs.uni-freiburg.de.
Authors: Muhammad Yaqub, Shahzad Ahmad, Malik Abdul Manan, Imran Shabir Chuhan
Abstract: Real-time traffic flow prediction holds significant importance within the domain of Intelligent Transportation Systems (ITS). The task of achieving a balance between prediction precision and computational efficiency presents a significant challenge. In this article, we present a novel deep-learning method called Federated Learning and Asynchronous Graph Convolutional Network (FLAGCN). Our framework incorporates the principles of asynchronous graph convolutional networks with federated learning to enhance the accuracy and efficiency of real-time traffic flow prediction. The FLAGCN model employs a spatial-temporal graph convolution technique to asynchronously address spatio-temporal dependencies within traffic data effectively. To efficiently handle the computational requirements associated with this deep learning model, this study used a graph federated learning technique known as GraphFL. This approach is designed to facilitate the training process. The experimental results obtained from conducting tests on two distinct traffic datasets demonstrate that the utilization of FLAGCN leads to the optimization of both training and inference durations while maintaining a high level of prediction accuracy. FLAGCN outperforms existing models with significant improvements by achieving up to approximately 6.85% reduction in RMSE, 20.45% reduction in MAPE, compared to the best-performing existing models.
Authors: Marien Renaud, Jean Prost, Arthur Leclaire, Nicolas Papadakis
Abstract: Plug-and-Play (PnP) algorithms are a class of iterative algorithms that address image inverse problems by combining a physical model and a deep neural network for regularization. Even if they produce impressive image restoration results, these algorithms rely on a non-standard use of a denoiser on images that are less and less noisy along the iterations, which contrasts with recent algorithms based on Diffusion Models (DM), where the denoiser is applied only on re-noised images. We propose a new PnP framework, called Stochastic deNOising REgularization (SNORE), which applies the denoiser only on images with noise of the adequate level. It is based on an explicit stochastic regularization, which leads to a stochastic gradient descent algorithm to solve ill-posed inverse problems. A convergence analysis of this algorithm and its annealing extension is provided. Experimentally, we prove that SNORE is competitive with respect to state-of-the-art methods on deblurring and inpainting tasks, both quantitatively and qualitatively.
Authors: Nate Gillman, Michael Freeman, Daksh Aggarwal, Chia-Hong Hsu, Calvin Luo, Yonglong Tian, Chen Sun
Abstract: As synthetic data becomes higher quality and proliferates on the internet, machine learning models are increasingly trained on a mix of human- and machine-generated data. Despite the successful stories of using synthetic data for representation learning, using synthetic data for generative model training creates "self-consuming loops" which may lead to training instability or even collapse, unless certain conditions are met. Our paper aims to stabilize self-consuming generative model training. Our theoretical results demonstrate that by introducing an idealized correction function, which maps a data point to be more likely under the true data distribution, self-consuming loops can be made exponentially more stable. We then propose self-correction functions, which rely on expert knowledge (e.g. the laws of physics programmed in a simulator), and aim to approximate the idealized corrector automatically and at scale. We empirically validate the effectiveness of self-correcting self-consuming loops on the challenging human motion synthesis task, and observe that it successfully avoids model collapse, even when the ratio of synthetic data to real data is as high as 100%.
Authors: Keller Jordan
Abstract: CIFAR-10 is among the most widely used datasets in machine learning, facilitating thousands of research projects per year. To accelerate research and reduce the cost of experiments, we introduce training methods for CIFAR-10 which reach 94% accuracy in 3.29 seconds, 95% in 10.4 seconds, and 96% in 46.3 seconds, when run on a single NVIDIA A100 GPU. As one factor contributing to these training speeds, we propose a derandomized variant of horizontal flipping augmentation, which we show improves over the standard method in every case where flipping is beneficial over no flipping at all. Our code is released at https://github.com/KellerJordan/cifar10-airbench.