Authors: Hidir Yesiltepe, Kiymet Akdemir, Pinar Yanardag
Abstract: Diffusion-based text-to-image models have rapidly gained popularity for their ability to generate detailed and realistic images from textual descriptions. However, these models often reflect the biases present in their training data, especially impacting marginalized groups. While prior efforts to debias language models have focused on addressing specific biases, such as racial or gender biases, efforts to tackle intersectional bias have been limited. Intersectional bias refers to the unique form of bias experienced by individuals at the intersection of multiple social identities. Addressing intersectional bias is crucial because it amplifies the negative effects of discrimination based on race, gender, and other identities. In this paper, we introduce a method that addresses intersectional bias in diffusion-based text-to-image models by modifying cross-attention maps in a disentangled manner. Our approach utilizes a pre-trained Stable Diffusion model, eliminates the need for an additional set of reference images, and preserves the original quality for unaltered concepts. Comprehensive experiments demonstrate that our method surpasses existing approaches in mitigating both single and intersectional biases across various attributes. We make our source code and debiased models for various attributes available to encourage fairness in generative models and to support further research.
Authors: Kevin Barkevich, Reynold Bailey, Gabriel J. Diaz
Abstract: Algorithms for the estimation of gaze direction from mobile and video-based eye trackers typically involve tracking a feature of the eye that moves through the eye camera image in a way that covaries with the shifting gaze direction, such as the center or boundaries of the pupil. Tracking these features using traditional computer vision techniques can be difficult due to partial occlusion and environmental reflections. Although recent efforts to use machine learning (ML) for pupil tracking have demonstrated superior results when evaluated using standard measures of segmentation performance, little is known of how these networks may affect the quality of the final gaze estimate. This work provides an objective assessment of the impact of several contemporary ML-based methods for eye feature tracking when the subsequent gaze estimate is produced using either feature-based or model-based methods. Metrics include the accuracy and precision of the gaze estimate, as well as drop-out rate.
Authors: Rolandos Alexandros Potamias, Michail Tarasiou Stylianos Ploumpis, Stefanos Zafeiriou
Abstract: In the realm of 3D computer vision, parametric models have emerged as a ground-breaking methodology for the creation of realistic and expressive 3D avatars. Traditionally, they rely on Principal Component Analysis (PCA), given its ability to decompose data to an orthonormal space that maximally captures shape variations. However, due to the orthogonality constraints and the global nature of PCA's decomposition, these models struggle to perform localized and disentangled editing of 3D shapes, which severely affects their use in applications requiring fine control such as face sculpting. In this paper, we leverage diffusion models to enable diverse and fully localized edits on 3D meshes, while completely preserving the un-edited regions. We propose an effective diffusion masking training strategy that, by design, facilitates localized manipulation of any shape region, without being limited to predefined regions or to sparse sets of predefined control vertices. Following our framework, a user can explicitly set their manipulation region of choice and define an arbitrary set of vertices as handles to edit a 3D mesh. Compared to the current state-of-the-art our method leads to more interpretable shape manipulations than methods relying on latent code state, greater localization and generation diversity while offering faster inference than optimization based approaches. Project page: https://rolpotamias.github.io/Shapefusion/
Authors: Tuna Han Salih Meral, Enis Simsar, Federico Tombari, Pinar Yanardag
Abstract: Low-Rank Adaptations (LoRAs) have emerged as a powerful and popular technique in the field of image generation, offering a highly effective way to adapt and refine pre-trained deep learning models for specific tasks without the need for comprehensive retraining. By employing pre-trained LoRA models, such as those representing a specific cat and a particular dog, the objective is to generate an image that faithfully embodies both animals as defined by the LoRAs. However, the task of seamlessly blending multiple concept LoRAs to capture a variety of concepts in one image proves to be a significant challenge. Common approaches often fall short, primarily because the attention mechanisms within different LoRA models overlap, leading to scenarios where one concept may be completely ignored (e.g., omitting the dog) or where concepts are incorrectly combined (e.g., producing an image of two cats instead of one cat and one dog). To overcome these issues, CLoRA addresses them by updating the attention maps of multiple LoRA models and leveraging them to create semantic masks that facilitate the fusion of latent representations. Our method enables the creation of composite images that truly reflect the characteristics of each LoRA, successfully merging multiple concepts or styles. Our comprehensive evaluations, both qualitative and quantitative, demonstrate that our approach outperforms existing methodologies, marking a significant advancement in the field of image generation with LoRAs. Furthermore, we share our source code, benchmark dataset, and trained LoRA models to promote further research on this topic.
Authors: Marco Cannici, Davide Scaramuzza
Abstract: Neural Radiance Fields (NeRFs) have shown great potential in novel view synthesis. However, they struggle to render sharp images when the data used for training is affected by motion blur. On the other hand, event cameras excel in dynamic scenes as they measure brightness changes with microsecond resolution and are thus only marginally affected by blur. Recent methods attempt to enhance NeRF reconstructions under camera motion by fusing frames and events. However, they face challenges in recovering accurate color content or constrain the NeRF to a set of predefined camera poses, harming reconstruction quality in challenging conditions. This paper proposes a novel formulation addressing these issues by leveraging both model- and learning-based modules. We explicitly model the blur formation process, exploiting the event double integral as an additional model-based prior. Additionally, we model the event-pixel response using an end-to-end learnable response function, allowing our method to adapt to non-idealities in the real event-camera sensor. We show, on synthetic and real data, that the proposed approach outperforms existing deblur NeRFs that use only frames as well as those that combine frames and events by +6.13dB and +2.48dB, respectively.
Authors: Seyed Rasoul Hosseini, Mohammad Teshnehlab
Abstract: Lane detection for autonomous vehicles is an important concept, yet it is a challenging issue of driver assistance systems in modern vehicles. The emergence of deep learning leads to significant progress in self-driving cars. Conventional deep learning-based methods handle lane detection problems as a binary segmentation task and determine whether a pixel belongs to a line. These methods rely on the assumption of a fixed number of lanes, which does not always work. This study aims to develop an optimal structure for the lane detection problem, offering a promising solution for driver assistance features in modern vehicles by utilizing a machine learning method consisting of binary segmentation and Affinity Fields that can manage varying numbers of lanes and lane change scenarios. In this approach, the Convolutional Neural Network (CNN), is selected as a feature extractor, and the final output is obtained through clustering of the semantic segmentation and Affinity Field outputs. Our method uses less complex CNN architecture than exi
Authors: Mingxing Rao, Yinhong Qin, Soheil Kolouri, Jie Ying Wu, Daniel Moyer
Abstract: Purpose: Surgical video is an important data stream for gesture recognition. Thus, robust visual encoders for those data-streams is similarly important. Methods: Leveraging the Bridge-Prompt framework, we fine-tune a pre-trained vision-text model (CLIP) for gesture recognition in surgical videos. This can utilize extensive outside video data such as text, but also make use of label meta-data and weakly supervised contrastive losses. Results: Our experiments show that prompt-based video encoder outperforms standard encoders in surgical gesture recognition tasks. Notably, it displays strong performance in zero-shot scenarios, where gestures/tasks that were not provided during the encoder training phase are included in the prediction phase. Additionally, we measure the benefit of inclusion text descriptions in the feature extractor training schema. Conclusion: Bridge-Prompt and similar pre-trained+fine-tuned video encoder models present significant visual representation for surgical robotics, especially in gesture recognition tasks. Given the diverse range of surgical tasks (gestures), the ability of these models to zero-shot transfer without the need for any task (gesture) specific retraining makes them invaluable.
Authors: Gabriele Berton, Gabriele Trivigno, Barbara Caputo, Carlo Masone
Abstract: Visual Place Recognition aims at recognizing previously visited places by relying on visual clues, and it is used in robotics applications for SLAM and localization. Since typically a mobile robot has access to a continuous stream of frames, this task is naturally cast as a sequence-to-sequence localization problem. Nevertheless, obtaining sequences of labelled data is much more expensive than collecting isolated images, which can be done in an automated way with little supervision. As a mitigation to this problem, we propose a novel Joint Image and Sequence Training protocol (JIST) that leverages large uncurated sets of images through a multi-task learning framework. With JIST we also introduce SeqGeM, an aggregation layer that revisits the popular GeM pooling to produce a single robust and compact embedding from a sequence of single-frame embeddings. We show that our model is able to outperform previous state of the art while being faster, using 8 times smaller descriptors, having a lighter architecture and allowing to process sequences of various lengths. Code is available at https://github.com/ga1i13o/JIST
Authors: George Tang, Krishna Murthy Jatavallabhula, Antonio Torralba
Abstract: We tackle the problem of learning an implicit scene representation for 3D instance segmentation from a sequence of posed RGB images. Towards this, we introduce 3DIML, a novel framework that efficiently learns a label field that may be rendered from novel viewpoints to produce view-consistent instance segmentation masks. 3DIML significantly improves upon training and inference runtimes of existing implicit scene representation based methods. Opposed to prior art that optimizes a neural field in a self-supervised manner, requiring complicated training procedures and loss function design, 3DIML leverages a two-phase process. The first phase, InstanceMap, takes as input 2D segmentation masks of the image sequence generated by a frontend instance segmentation model, and associates corresponding masks across images to 3D labels. These almost view-consistent pseudolabel masks are then used in the second phase, InstanceLift, to supervise the training of a neural label field, which interpolates regions missed by InstanceMap and resolves ambiguities. Additionally, we introduce InstanceLoc, which enables near realtime localization of instance masks given a trained label field and an off-the-shelf image segmentation model by fusing outputs from both. We evaluate 3DIML on sequences from the Replica and ScanNet datasets and demonstrate 3DIML's effectiveness under mild assumptions for the image sequences. We achieve a 14-24x speedup over existing implicit scene representation methods with comparable quality, showcasing its potential to facilitate faster and more effective 3D scene understanding.
Authors: Anna Kukleva, Fadime Sener, Edoardo Remelli, Bugra Tekin, Eric Sauser, Bernt Schiele, Shugao Ma
Abstract: Lately, there has been growing interest in adapting vision-language models (VLMs) to image and third-person video classification due to their success in zero-shot recognition. However, the adaptation of these models to egocentric videos has been largely unexplored. To address this gap, we propose a simple yet effective cross-modal adaptation framework, which we call X-MIC. Using a video adapter, our pipeline learns to align frozen text embeddings to each egocentric video directly in the shared embedding space. Our novel adapter architecture retains and improves generalization of the pre-trained VLMs by disentangling learnable temporal modeling and frozen visual encoder. This results in an enhanced alignment of text embeddings to each egocentric video, leading to a significant improvement in cross-dataset generalization. We evaluate our approach on the Epic-Kitchens, Ego4D, and EGTEA datasets for fine-grained cross-dataset action generalization, demonstrating the effectiveness of our method. Code is available at https://github.com/annusha/xmic
Authors: Akshay Gopalkrishnan, Ross Greer, Mohan Trivedi
Abstract: Vision-Language Models (VLMs) and Multi-Modal Language models (MMLMs) have become prominent in autonomous driving research, as these models can provide interpretable textual reasoning and responses for end-to-end autonomous driving safety tasks using traffic scene images and other data modalities. However, current approaches to these systems use expensive large language model (LLM) backbones and image encoders, making such systems unsuitable for real-time autonomous driving systems where tight memory constraints exist and fast inference time is necessary. To address these previous issues, we develop EM-VLM4AD, an efficient, lightweight, multi-frame vision language model which performs Visual Question Answering for autonomous driving. In comparison to previous approaches, EM-VLM4AD requires at least 10 times less memory and floating point operations, while also achieving higher BLEU-4, METEOR, CIDEr, and ROGUE scores than the existing baseline on the DriveLM dataset. EM-VLM4AD also exhibits the ability to extract relevant information from traffic views related to prompts and can answer questions for various autonomous driving subtasks. We release our code to train and evaluate our model at https://github.com/akshaygopalkr/EM-VLM4AD.
Authors: Yuhang Li, Xin Dong, Chen Chen, Jingtao Li, Yuxin Wen, Michael Spranger, Lingjuan Lyu
Abstract: Synthetic image data generation represents a promising avenue for training deep learning models, particularly in the realm of transfer learning, where obtaining real images within a specific domain can be prohibitively expensive due to privacy and intellectual property considerations. This work delves into the generation and utilization of synthetic images derived from text-to-image generative models in facilitating transfer learning paradigms. Despite the high visual fidelity of the generated images, we observe that their naive incorporation into existing real-image datasets does not consistently enhance model performance due to the inherent distribution gap between synthetic and real images. To address this issue, we introduce a novel two-stage framework called bridged transfer, which initially employs synthetic images for fine-tuning a pre-trained model to improve its transferability and subsequently uses real data for rapid adaptation. Alongside, We propose dataset style inversion strategy to improve the stylistic alignment between synthetic and real images. Our proposed methods are evaluated across 10 different datasets and 5 distinct models, demonstrating consistent improvements, with up to 30% accuracy increase on classification tasks. Intriguingly, we note that the enhancements were not yet saturated, indicating that the benefits may further increase with an expanded volume of synthetic data.
Authors: Colin Keil, Aniket Gupta, Pushyami Kaveti, Hanumant Singh
Abstract: Visual SLAM with thermal imagery, and other low contrast visually degraded environments such as underwater, or in areas dominated by snow and ice, remain a difficult problem for many state of the art (SOTA) algorithms. In addition to challenging front-end data association, thermal imagery presents an additional difficulty for long term relocalization and map reuse. The relative temperatures of objects in thermal imagery change dramatically from day to night. Feature descriptors typically used for relocalization in SLAM are unable to maintain consistency over these diurnal changes. We show that learned feature descriptors can be used within existing Bag of Word based localization schemes to dramatically improve place recognition across large temporal gaps in thermal imagery. In order to demonstrate the effectiveness of our trained vocabulary, we have developed a baseline SLAM system, integrating learned features and matching into a classical SLAM algorithm. Our system demonstrates good local tracking on challenging thermal imagery, and relocalization that overcomes dramatic day to night thermal appearance changes. Our code and datasets are available here: https://github.com/neufieldrobotics/IRSLAM_Baseline
Authors: Ruining Yang, Yuqi Peng
Abstract: Autonomous driving has garnered significant attention as a key research area within artificial intelligence. In the context of autonomous driving scenarios, the varying physical locations of objects correspond to different levels of danger. However, conventional evaluation criteria for automatic driving object detection often overlook the crucial aspect of an object's physical location, leading to evaluation results that may not accurately reflect the genuine threat posed by the object to the autonomous driving vehicle. To enhance the safety of autonomous driving, this paper introduces a novel evaluation criterion based on physical location information, termed PLoc. This criterion transcends the limitations of traditional criteria by acknowledging that the physical location of pedestrians in autonomous driving scenarios can provide valuable safety-related information. Furthermore, this paper presents a newly re-annotated dataset (ApolloScape-R) derived from ApolloScape. ApolloScape-R involves the relabeling of pedestrians based on the significance of their physical location. The dataset is utilized to assess the performance of various object detection models under the proposed PLoc criterion. Experimental results demonstrate that the average accuracy of all object detection models in identifying a person situated in the travel lane of an autonomous vehicle is lower than that for a person on a sidewalk. The dataset is publicly available at https://github.com/lnyrlyed/ApolloScape-R.git
Authors: Seyma Yucer, Amir Atapour Abarghouei, Noura Al Moubayed, Toby P. Breckon
Abstract: Achieving an effective fine-grained appearance variation over 2D facial images, whilst preserving facial identity, is a challenging task due to the high complexity and entanglement of common 2D facial feature encoding spaces. Despite these challenges, such fine-grained control, by way of disentanglement is a crucial enabler for data-driven racial bias mitigation strategies across multiple automated facial analysis tasks, as it allows to analyse, characterise and synthesise human facial diversity. In this paper, we propose a novel GAN framework to enable fine-grained control over individual race-related phenotype attributes of the facial images. Our framework factors the latent (feature) space into elements that correspond to race-related facial phenotype representations, thereby separating phenotype aspects (e.g. skin, hair colour, nose, eye, mouth shapes), which are notoriously difficult to annotate robustly in real-world facial data. Concurrently, we also introduce a high quality augmented, diverse 2D face image dataset drawn from CelebA-HQ for GAN training. Unlike prior work, our framework only relies upon 2D imagery and related parameters to achieve state-of-the-art individual control over race-related phenotype attributes with improved photo-realistic output.
Authors: Haipeng Liu, Yang Wang, Biao Qian, Meng Wang, Yong Rui
Abstract: Denoising diffusion probabilistic models for image inpainting aim to add the noise to the texture of image during the forward process and recover masked regions with unmasked ones of the texture via the reverse denoising process.Despite the meaningful semantics generation,the existing arts suffer from the semantic discrepancy between masked and unmasked regions, since the semantically dense unmasked texture fails to be completely degraded while the masked regions turn to the pure noise in diffusion process,leading to the large discrepancy between them.In this paper,we aim to answer how unmasked semantics guide texture denoising process;together with how to tackle the semantic discrepancy,to facilitate the consistent and meaningful semantics generation.To this end,we propose a novel structure-guided diffusion model named StrDiffusion,to reformulate the conventional texture denoising process under structure guidance to derive a simplified denoising objective for image inpainting,while revealing:1) the semantically sparse structure is beneficial to tackle semantic discrepancy in early stage, while dense texture generates reasonable semantics in late stage;2) the semantics from unmasked regions essentially offer the time-dependent structure guidance for the texture denoising process,benefiting from the time-dependent sparsity of the structure semantics.For the denoising process,a structure-guided neural network is trained to estimate the simplified denoising objective by exploiting the consistency of the denoised structure between masked and unmasked regions.Besides,we devise an adaptive resampling strategy as a formal criterion as whether structure is competent to guide the texture denoising process,while regulate their semantic correlations.Extensive experiments validate the merits of StrDiffusion over the state-of-the-arts.Our code is available at https://github.com/htyjers/StrDiffusion.
Authors: Jianfeng Cai, Yue Ma, Zhixi Feng, Shuyuan Yang
Abstract: Polarimetric synthetic aperture radar (PolSAR) image interpretation is widely used in various fields. Recently, deep learning has made significant progress in PolSAR image classification. Supervised learning (SL) requires a large amount of labeled PolSAR data with high quality to achieve better performance, however, manually labeled data is insufficient. This causes the SL to fail into overfitting and degrades its generalization performance. Furthermore, the scattering confusion problem is also a significant challenge that attracts more attention. To solve these problems, this article proposes a Heterogeneous Network based Contrastive Learning method(HCLNet). It aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multi-features and superpixels. Beyond the conventional CL, HCLNet introduces the heterogeneous architecture for the first time to utilize heterogeneous PolSAR features better. And it develops two easy-to-use plugins to narrow the domain gap between optics and PolSAR, including feature filter and superpixel-based instance discrimination, which the former is used to enhance the complementarity of multi-features, and the latter is used to increase the diversity of negative samples. Experiments demonstrate the superiority of HCLNet on three widely used PolSAR benchmark datasets compared with state-of-the-art methods. Ablation studies also verify the importance of each component. Besides, this work has implications for how to efficiently utilize the multi-features of PolSAR data to learn better high-level representation in CL and how to construct networks suitable for PolSAR data better.
Authors: Junho Kim, Jiwon Jeong, Young Min Kim
Abstract: We introduce a lightweight and accurate localization method that only utilizes the geometry of 2D-3D lines. Given a pre-captured 3D map, our approach localizes a panorama image, taking advantage of the holistic 360 view. The system mitigates potential privacy breaches or domain discrepancies by avoiding trained or hand-crafted visual descriptors. However, as lines alone can be ambiguous, we express distinctive yet compact spatial contexts from relationships between lines, namely the dominant directions of parallel lines and the intersection between non-parallel lines. The resulting representations are efficient in processing time and memory compared to conventional visual descriptor-based methods. Given the groups of dominant line directions and their intersections, we accelerate the search process to test thousands of pose candidates in less than a millisecond without sacrificing accuracy. We empirically show that the proposed 2D-3D matching can localize panoramas for challenging scenes with similar structures, dramatic domain shifts or illumination changes. Our fully geometric approach does not involve extensive parameter tuning or neural network training, making it a practical algorithm that can be readily deployed in the real world. Project page including the code is available through this link: https://82magnolia.github.io/fgpl/.
Authors: Inas Al-Kamachy (Karlstad University, Sweden), Prof. Dr. Reza Hassanpour (Rotterdam University, Netherlands), Prof. Roya Choupani (Angelo State University, USA)
Abstract: Diabetic Retinopathy (DR) stands as the leading cause of blindness globally, particularly affecting individuals between the ages of 20 and 70. This paper presents a Computer-Aided Diagnosis (CAD) system designed for the automatic classification of retinal images into five distinct classes: Normal, Mild, Moderate, Severe, and Proliferative Diabetic Retinopathy (PDR). The proposed system leverages Convolutional Neural Networks (CNNs) employing pre-trained deep learning models. Through the application of fine-tuning techniques, our model is trained on fundus images of diabetic retinopathy with resolutions of 350x350x3 and 224x224x3. Experimental results obtained on the Kaggle platform, utilizing resources comprising 4 CPUs, 17 GB RAM, and 1 GB Disk, demonstrate the efficacy of our approach. The achieved Area Under the Curve (AUC) values for CNN, MobileNet, VGG-16, InceptionV3, and InceptionResNetV2 models are 0.50, 0.70, 0.53, 0.63, and 0.69, respectively.
Authors: Zihao Song, Huaxi Chen, Donghui Quan, Di Li, Yinghui Zheng, Shulei Ni, Yunchuan Chen, Yun Zheng
Abstract: We introduce a machine learning-based method for extracting HI sources from 3D spectral data, and construct a dedicated dataset of HI sources from CRAFTS. Our custom dataset provides comprehensive resources for HI source detection. Utilizing the 3D-Unet segmentation architecture, our method reliably identifies and segments HI sources, achieving notable performance metrics with recall rates reaching 91.6% and accuracy levels at 95.7%. These outcomes substantiate the value of our custom dataset and the efficacy of our proposed network in identifying HI source. Our code is publicly available at https://github.com/fishszh/HISF.
Authors: Qianliang Wu, Haobo Jiang, Lei Luo, Jun Li, Yaqing Ding, Jin Xie, Jian Yang
Abstract: Establishing reliable correspondences is essential for registration tasks such as 3D and 2D3D registration. Existing methods commonly leverage geometric or semantic point features to generate potential correspondences. However, these features may face challenges such as large deformation, scale inconsistency, and ambiguous matching problems (e.g., symmetry). Additionally, many previous methods, which rely on single-pass prediction, may struggle with local minima in complex scenarios. To mitigate these challenges, we introduce a diffusion matching model for robust correspondence construction. Our approach treats correspondence estimation as a denoising diffusion process within the doubly stochastic matrix space, which gradually denoises (refines) a doubly stochastic matching matrix to the ground-truth one for high-quality correspondence estimation. It involves a forward diffusion process that gradually introduces Gaussian noise into the ground truth matching matrix and a reverse denoising process that iteratively refines the noisy matching matrix. In particular, the feature extraction from the backbone occurs only once during the inference phase. Our lightweight denoising module utilizes the same feature at each reverse sampling step. Evaluation of our method on both 3D and 2D3D registration tasks confirms its effectiveness.
Authors: Aggelina Chatziagapi, Grigorios G. Chrysos, Dimitris Samaras
Abstract: In this work, we introduce a method that learns a single dynamic neural radiance field (NeRF) from monocular talking face videos of multiple identities. NeRFs have shown remarkable results in modeling the 4D dynamics and appearance of human faces. However, they require per-identity optimization. Although recent approaches have proposed techniques to reduce the training and rendering time, increasing the number of identities can be expensive. We introduce MI-NeRF (multi-identity NeRF), a single unified network that models complex non-rigid facial motion for multiple identities, using only monocular videos of arbitrary length. The core premise in our method is to learn the non-linear interactions between identity and non-identity specific information with a multiplicative module. By training on multiple videos simultaneously, MI-NeRF not only reduces the total training time compared to standard single-identity NeRFs, but also demonstrates robustness in synthesizing novel expressions for any input identity. We present results for both facial expression transfer and talking face video synthesis. Our method can be further personalized for a target identity given only a short video.
Authors: Bo Wang, Jian Li, Yang Yu, Li Liu, Zhenping Sun, Dewen Hu
Abstract: Considering the complementarity of scene flow estimation in the spatial domain's focusing capability and 3D object tracking in the temporal domain's coherence, this study aims to address a comprehensive new task that can simultaneously capture fine-grained and long-term 3D motion in an online manner: long-term scene flow estimation (LSFE). We introduce SceneTracker, a novel learning-based LSFE network that adopts an iterative approach to approximate the optimal trajectory. Besides, it dynamically indexes and constructs appearance and depth correlation features simultaneously and employs the Transformer to explore and utilize long-range connections within and between trajectories. With detailed experiments, SceneTracker shows superior capabilities in handling 3D spatial occlusion and depth noise interference, highly tailored to the LSFE task's needs. The code for SceneTracker is available at https://github.com/wwsource/SceneTracker.
Authors: Jijie He, Wenwu Yang
Abstract: By leveraging temporal dependency in video sequences, multi-frame human pose estimation algorithms have demonstrated remarkable results in complicated situations, such as occlusion, motion blur, and video defocus. These algorithms are predominantly based on heatmaps, resulting in high computation and storage requirements per frame, which limits their flexibility and real-time application in video scenarios, particularly on edge devices. In this paper, we develop an efficient and effective video-based human pose regression method, which bypasses intermediate representations such as heatmaps and instead directly maps the input to the output joint coordinates. Despite the inherent spatial correlation among adjacent joints of the human pose, the temporal trajectory of each individual joint exhibits relative independence. In light of this, we propose a novel Decoupled Space-Time Aggregation network (DSTA) to separately capture the spatial contexts between adjacent joints and the temporal cues of each individual joint, thereby avoiding the conflation of spatiotemporal dimensions. Concretely, DSTA learns a dedicated feature token for each joint to facilitate the modeling of their spatiotemporal dependencies. With the proposed joint-wise local-awareness attention mechanism, our method is capable of efficiently and flexibly utilizing the spatial dependency of adjacent joints and the temporal dependency of each joint itself. Extensive experiments demonstrate the superiority of our method. Compared to previous regression-based single-frame human pose estimation methods, DSTA significantly enhances performance, achieving an 8.9 mAP improvement on PoseTrack2017. Furthermore, our approach either surpasses or is on par with the state-of-the-art heatmap-based multi-frame human pose estimation methods. Project page: https://github.com/zgspose/DSTA.
Authors: Artur Santos Nascimento, Valter Guilherme Silva de Souza, Daniel Oliveira Dantas, Beatriz Trinch\~ao Andrade
Abstract: In computer vision, characteristics refer to image regions with unique properties, such as corners, edges, textures, or areas with high contrast. These regions can be represented through feature points (FPs). FP detection and description are fundamental steps to many computer vision tasks. Most FP detection and description methods use low dynamic range (LDR) images, sufficient for most applications involving digital images. However, LDR images may have saturated pixels in scenes with extreme light conditions, which degrade FP detection. On the other hand, high dynamic range (HDR) images usually present a greater dynamic range but FP detection algorithms do not take advantage of all the information in such images. In this study, we present a systematic review of image detection and description algorithms that use HDR images as input. We developed a library called CP_HDR that implements the Harris corner detector, SIFT detector and descriptor, and two modifications of those algorithms specialized in HDR images, called SIFT for HDR (SfHDR) and Harris for HDR (HfHDR). Previous studies investigated the use of HDR images in FP detection, but we did not find studies investigating the use of HDR images in FP description. Using uniformity, repeatability rate, mean average precision, and matching rate metrics, we compared the performance of the CP_HDR algorithms using LDR and HDR images. We observed an increase in the uniformity of the distribution of FPs among the high-light, mid-light, and low-light areas of the images. The results show that using HDR images as input to detection algorithms improves performance and that SfHDR and HfHDR enhance FP description.
Authors: Gengchen Zhang, Yulun Zhang, Xin Yuan, Ying Fu
Abstract: Recently, deep neural networks have achieved excellent performance on low-light raw video enhancement. However, they often come with high computational complexity and large memory costs, which hinder their applications on resource-limited devices. In this paper, we explore the feasibility of applying the extremely compact binary neural network (BNN) to low-light raw video enhancement. Nevertheless, there are two main issues with binarizing video enhancement models. One is how to fuse the temporal information to improve low-light denoising without complex modules. The other is how to narrow the performance gap between binary convolutions with the full precision ones. To address the first issue, we introduce a spatial-temporal shift operation, which is easy-to-binarize and effective. The temporal shift efficiently aggregates the features of neighbor frames and the spatial shift handles the misalignment caused by the large motion in videos. For the second issue, we present a distribution-aware binary convolution, which captures the distribution characteristics of real-valued input and incorporates them into plain binary convolutions to alleviate the degradation in performance. Extensive quantitative and qualitative experiments have shown our high-efficiency binarized low-light raw video enhancement method can attain a promising performance.
Authors: Yan Luo, Min Shi, Muhammad Osama Khan, Muhammad Muneeb Afzal, Hao Huang, Shuaihang Yuan, Yu Tian, Luo Song, Ava Kouhana, Tobias Elze, Yi Fang, Mengyu Wang
Abstract: Fairness is a critical concern in deep learning, especially in healthcare, where these models influence diagnoses and treatment decisions. Although fairness has been investigated in the vision-only domain, the fairness of medical vision-language (VL) models remains unexplored due to the scarcity of medical VL datasets for studying fairness. To bridge this research gap, we introduce the first fair vision-language medical dataset FairVLMed that provides detailed demographic attributes, ground-truth labels, and clinical notes to facilitate an in-depth examination of fairness within VL foundation models. Using FairVLMed, we conduct a comprehensive fairness analysis of two widely-used VL models (CLIP and BLIP2), pre-trained on both natural and medical domains, across four different protected attributes. Our results highlight significant biases in all VL models, with Asian, Male, Non-Hispanic, and Spanish being the preferred subgroups across the protected attributes of race, gender, ethnicity, and language, respectively. In order to alleviate these biases, we propose FairCLIP, an optimal-transport-based approach that achieves a favorable trade-off between performance and fairness by reducing the Sinkhorn distance between the overall sample distribution and the distributions corresponding to each demographic group. As the first VL dataset of its kind, FairVLMed holds the potential to catalyze advancements in the development of machine learning models that are both ethically aware and clinically effective. Our dataset and code are available at https://ophai.hms.harvard.edu/datasets/fairvlmed10k.
Authors: Xu Ma, Xiyang Dai, Jianwei Yang, Bin Xiao, Yinpeng Chen, Yun Fu, Lu Yuan
Abstract: In this work, we present efficient modulation, a novel design for efficient vision networks. We revisit the modulation mechanism, which operates input through convolutional context modeling and feature projection layers, and fuses features via element-wise multiplication and an MLP block. We demonstrate that the modulation mechanism is particularly well suited for efficient networks and further tailor the modulation design by proposing the efficient modulation (EfficientMod) block, which is considered the essential building block for our networks. Benefiting from the prominent representational ability of modulation mechanism and the proposed efficient design, our network can accomplish better trade-offs between accuracy and efficiency and set new state-of-the-art performance in the zoo of efficient networks. When integrating EfficientMod with the vanilla self-attention block, we obtain the hybrid architecture which further improves the performance without loss of efficiency. We carry out comprehensive experiments to verify EfficientMod's performance. With fewer parameters, our EfficientMod-s performs 0.6 top-1 accuracy better than EfficientFormerV2-s2 and is 25% faster on GPU, and 2.9 better than MobileViTv2-1.0 at the same GPU latency. Additionally, our method presents a notable improvement in downstream tasks, outperforming EfficientFormerV2-s by 3.6 mIoU on the ADE20K benchmark. Code and checkpoints are available at https://github.com/ma-xu/EfficientMod.
Authors: Robik Shrestha, Yang Zou, Qiuyu Chen, Zhiheng Li, Yusheng Xie, Siqi Deng
Abstract: Existing text-to-image generative models reflect or even amplify societal biases ingrained in their training data. This is especially concerning for human image generation where models are biased against certain demographic groups. Existing attempts to rectify this issue are hindered by the inherent limitations of the pre-trained models and fail to substantially improve demographic diversity. In this work, we introduce Fair Retrieval Augmented Generation (FairRAG), a novel framework that conditions pre-trained generative models on reference images retrieved from an external image database to improve fairness in human generation. FairRAG enables conditioning through a lightweight linear module that projects reference images into the textual space. To enhance fairness, FairRAG applies simple-yet-effective debiasing strategies, providing images from diverse demographic groups during the generative process. Extensive experiments demonstrate that FairRAG outperforms existing methods in terms of demographic diversity, image-text alignment, and image fidelity while incurring minimal computational overhead during inference.
Authors: Xu Ma, Xiyang Dai, Yue Bai, Yizhou Wang, Yun Fu
Abstract: Recent studies have drawn attention to the untapped potential of the "star operation" (element-wise multiplication) in network design. While intuitive explanations abound, the foundational rationale behind its application remains largely unexplored. Our study attempts to reveal the star operation's ability to map inputs into high-dimensional, non-linear feature spaces -- akin to kernel tricks -- without widening the network. We further introduce StarNet, a simple yet powerful prototype, demonstrating impressive performance and low latency under compact network structure and efficient budget. Like stars in the sky, the star operation appears unremarkable but holds a vast universe of potential. Our work encourages further exploration across tasks, with codes available at https://github.com/ma-xu/Rewrite-the-Stars.
Authors: Guanhua Ding, Zexi Ye, Zhen Zhong, Gang Li, David Shao
Abstract: Deep Neural Network (DNN) pruning has emerged as a key strategy to reduce model size, improve inference latency, and lower power consumption on DNN accelerators. Among various pruning techniques, block and output channel pruning have shown significant potential in accelerating hardware performance. However, their accuracy often requires further improvement. In response to this challenge, we introduce a separate, dynamic and differentiable (SMART) pruner. This pruner stands out by utilizing a separate, learnable probability mask for weight importance ranking, employing a differentiable Top k operator to achieve target sparsity, and leveraging a dynamic temperature parameter trick to escape from non-sparse local minima. In our experiments, the SMART pruner consistently demonstrated its superiority over existing pruning methods across a wide range of tasks and models on block and output channel pruning. Additionally, we extend our testing to Transformer-based models in N:M pruning scenarios, where SMART pruner also yields state-of-the-art results, demonstrating its adaptability and robustness across various neural network architectures, and pruning types.
Authors: Yanyan Shao, Shuting He, Qi Ye, Yuchao Feng, Wenhan Luo, Jiming Chen
Abstract: Tracking by natural language specification (TNL) aims to consistently localize a target in a video sequence given a linguistic description in the initial frame. Existing methodologies perform language-based and template-based matching for target reasoning separately and merge the matching results from two sources, which suffer from tracking drift when language and visual templates miss-align with the dynamic target state and ambiguity in the later merging stage. To tackle the issues, we propose a joint multi-modal tracking framework with 1) a prompt modulation module to leverage the complementarity between temporal visual templates and language expressions, enabling precise and context-aware appearance and linguistic cues, and 2) a unified target decoding module to integrate the multi-modal reference cues and executes the integrated queries on the search image to predict the target location in an end-to-end manner directly. This design ensures spatio-temporal consistency by leveraging historical visual information and introduces an integrated solution, generating predictions in a single step. Extensive experiments conducted on TNL2K, OTB-Lang, LaSOT, and RefCOCOg validate the efficacy of our proposed approach. The results demonstrate competitive performance against state-of-the-art methods for both tracking and grounding.
Authors: Aayush Atul Verma, Bharatesh Chakravarthi, Arpitsinh Vaghela, Hua Wei, Yezhou Yang
Abstract: Event cameras, with their high temporal and dynamic range and minimal memory usage, have found applications in various fields. However, their potential in static traffic monitoring remains largely unexplored. To facilitate this exploration, we present eTraM - a first-of-its-kind, fully event-based traffic monitoring dataset. eTraM offers 10 hr of data from different traffic scenarios in various lighting and weather conditions, providing a comprehensive overview of real-world situations. Providing 2M bounding box annotations, it covers eight distinct classes of traffic participants, ranging from vehicles to pedestrians and micro-mobility. eTraM's utility has been assessed using state-of-the-art methods for traffic participant detection, including RVT, RED, and YOLOv8. We quantitatively evaluate the ability of event-based models to generalize on nighttime and unseen scenes. Our findings substantiate the compelling potential of leveraging event cameras for traffic monitoring, opening new avenues for research and application. eTraM is available at https://eventbasedvision.github.io/eTraM
Authors: Yuwen Tan, Qinhao Zhou, Xiang Xiang, Ke Wang, Yuchuan Wu, Yongbin Li
Abstract: Class-incremental learning (CIL) aims to enable models to continuously learn new classes while overcoming catastrophic forgetting. The introduction of pre-trained models has brought new tuning paradigms to CIL. In this paper, we revisit different parameter-efficient tuning (PET) methods within the context of continual learning. We observe that adapter tuning demonstrates superiority over prompt-based methods, even without parameter expansion in each learning session. Motivated by this, we propose incrementally tuning the shared adapter without imposing parameter update constraints, enhancing the learning capacity of the backbone. Additionally, we employ feature sampling from stored prototypes to retrain a unified classifier, further improving its performance. We estimate the semantic shift of old prototypes without access to past samples and update stored prototypes session by session. Our proposed method eliminates model expansion and avoids retaining any image samples. It surpasses previous pre-trained model-based CIL methods and demonstrates remarkable continual learning capabilities. Experimental results on five CIL benchmarks validate the effectiveness of our approach, achieving state-of-the-art (SOTA) performance.
Authors: Jiayu Li, Xuechao Zou, Shiying Wang, Ben Chen, Junliang Xing, Pin Tao
Abstract: Cattle face recognition holds paramount significance in domains such as animal husbandry and behavioral research. Despite significant progress in confined environments, applying these accomplishments in wild settings remains challenging. Thus, we create the first large-scale cattle face recognition dataset, ICRWE, for wild environments. It encompasses 483 cattle and 9,816 high-resolution image samples. Each sample undergoes annotation for face features, light conditions, and face orientation. Furthermore, we introduce a novel parallel attention network, PANet. Comprising several cascaded Transformer modules, each module incorporates two parallel Position Attention Modules (PAM) and Feature Mapping Modules (FMM). PAM focuses on local and global features at each image position through parallel channel attention, and FMM captures intricate feature patterns through non-linear mappings. Experimental results indicate that PANet achieves a recognition accuracy of 88.03% on the ICRWE dataset, establishing itself as the current state-of-the-art approach. The source code is available in the supplementary materials.
Authors: Byeongin Joung, Byeong-Uk Lee, Jaesung Choe, Ukcheol Shin, Minjun Kang, Taeyeop Lee, In So Kweon, Kuk-Jin Yoon
Abstract: This paper proposes an algorithm for synthesizing novel views under few-shot setup. The main concept is to develop a stable surface regularization technique called Annealing Signed Distance Function (ASDF), which anneals the surface in a coarse-to-fine manner to accelerate convergence speed. We observe that the Eikonal loss - which is a widely known geometric regularization - requires dense training signal to shape different level-sets of SDF, leading to low-fidelity results under few-shot training. In contrast, the proposed surface regularization successfully reconstructs scenes and produce high-fidelity geometry with stable training. Our method is further accelerated by utilizing grid representation and monocular geometric priors. Finally, the proposed approach is up to 45 times faster than existing few-shot novel view synthesis methods, and it produces comparable results in the ScanNet dataset and NeRF-Real dataset.
Authors: Zelin Zhao, Fenglei Fan, Wenlong Liao, Junchi Yan
Abstract: Many contemporary studies utilize grid-based models for neural field representation, but a systematic analysis of grid-based models is still missing, hindering the improvement of those models. Therefore, this paper introduces a theoretical framework for grid-based models. This framework points out that these models' approximation and generalization behaviors are determined by grid tangent kernels (GTK), which are intrinsic properties of grid-based models. The proposed framework facilitates a consistent and systematic analysis of diverse grid-based models. Furthermore, the introduced framework motivates the development of a novel grid-based model named the Multiplicative Fourier Adaptive Grid (MulFAGrid). The numerical analysis demonstrates that MulFAGrid exhibits a lower generalization bound than its predecessors, indicating its robust generalization performance. Empirical studies reveal that MulFAGrid achieves state-of-the-art performance in various tasks, including 2D image fitting, 3D signed distance field (SDF) reconstruction, and novel view synthesis, demonstrating superior representation ability. The project website is available at https://sites.google.com/view/cvpr24-2034-submission/home.
URLs: https://sites.google.com/view/cvpr24-2034-submission/home.
Authors: Juhwan Choi, YoungBin Kim
Abstract: Data augmentation is one of the regularization strategies for the training of deep learning models, which enhances generalizability and prevents overfitting, leading to performance improvement. Although researchers have proposed various data augmentation techniques, they often lack consideration for the difficulty of augmented data. Recently, another line of research suggests incorporating the concept of curriculum learning with data augmentation in the field of natural language processing. In this study, we adopt curriculum data augmentation for image data augmentation and propose colorful cutout, which gradually increases the noise and difficulty introduced in the augmented image. Our experimental results highlight the possibility of curriculum data augmentation for image data. We publicly released our source code to improve the reproducibility of our study.
Authors: Yunhao Li, Jing Wu, Lingzhe Zhao, Peidong Liu
Abstract: When capturing images through the glass during rainy or snowy weather conditions, the resulting images often contain waterdrops adhered on the glass surface, and these waterdrops significantly degrade the image quality and performance of many computer vision algorithms. To tackle these limitations, we propose a method to reconstruct the clear 3D scene implicitly from multi-view images degraded by waterdrops. Our method exploits an attention network to predict the location of waterdrops and then train a Neural Radiance Fields to recover the 3D scene implicitly. By leveraging the strong scene representation capabilities of NeRF, our method can render high-quality novel-view images with waterdrops removed. Extensive experimental results on both synthetic and real datasets show that our method is able to generate clear 3D scenes and outperforms existing state-of-the-art (SOTA) image adhesive waterdrop removal methods.
Authors: Ruijie Quan, Wenguan Wang, Zhibo Tian, Fan Ma, Yi Yang
Abstract: Reconstructing the viewed images from human brain activity bridges human and computer vision through the Brain-Computer Interface. The inherent variability in brain function between individuals leads existing literature to focus on acquiring separate models for each individual using their respective brain signal data, ignoring commonalities between these data. In this article, we devise Psychometry, an omnifit model for reconstructing images from functional Magnetic Resonance Imaging (fMRI) obtained from different subjects. Psychometry incorporates an omni mixture-of-experts (Omni MoE) module where all the experts work together to capture the inter-subject commonalities, while each expert associated with subject-specific parameters copes with the individual differences. Moreover, Psychometry is equipped with a retrieval-enhanced inference strategy, termed Ecphory, which aims to enhance the learned fMRI representation via retrieving from prestored subject-specific memories. These designs collectively render Psychometry omnifit and efficient, enabling it to capture both inter-subject commonality and individual specificity across subjects. As a result, the enhanced fMRI representations serve as conditional signals to guide a generation model to reconstruct high-quality and realistic images, establishing Psychometry as state-of-the-art in terms of both high-level and low-level metrics.
Authors: Shuang Li, Jiahua Wang, Lijie Wen
Abstract: Multi-modal reasoning plays a vital role in bridging the gap between textual and visual information, enabling a deeper understanding of the context. This paper presents the Feature Swapping Multi-modal Reasoning (FSMR) model, designed to enhance multi-modal reasoning through feature swapping. FSMR leverages a pre-trained visual-language model as an encoder, accommodating both text and image inputs for effective feature representation from both modalities. It introduces a unique feature swapping module, enabling the exchange of features between identified objects in images and corresponding vocabulary words in text, thereby enhancing the model's comprehension of the interplay between images and text. To further bolster its multi-modal alignment capabilities, FSMR incorporates a multi-modal cross-attention mechanism, facilitating the joint modeling of textual and visual information. During training, we employ image-text matching and cross-entropy losses to ensure semantic consistency between visual and language elements. Extensive experiments on the PMR dataset demonstrate FSMR's superiority over state-of-the-art baseline models across various performance metrics.
Authors: Yiteng Xu, Kecheng Ye, Xiao Han, Yiming Ren, Xinge Zhu, Yuexin Ma
Abstract: Human-centric Point Cloud Video Understanding (PVU) is an emerging field focused on extracting and interpreting human-related features from sequences of human point clouds, further advancing downstream human-centric tasks and applications. Previous works usually focus on tackling one specific task and rely on huge labeled data, which has poor generalization capability. Considering that human has specific characteristics, including the structural semantics of human body and the dynamics of human motions, we propose a unified framework to make full use of the prior knowledge and explore the inherent features in the data itself for generalized human-centric point cloud video understanding. Extensive experiments demonstrate that our method achieves state-of-the-art performance on various human-related tasks, including action recognition and 3D pose estimation. All datasets and code will be released soon.
Authors: Zhuopeng Li, Yilin Zhang, Chenming Wu, Jianke Zhu, Liangjun Zhang
Abstract: The rapid growth of 3D Gaussian Splatting (3DGS) has revolutionized neural rendering, enabling real-time production of high-quality renderings. However, the previous 3DGS-based methods have limitations in urban scenes due to reliance on initial Structure-from-Motion(SfM) points and difficulties in rendering distant, sky and low-texture areas. To overcome these challenges, we propose a hybrid optimization method named HO-Gaussian, which combines a grid-based volume with the 3DGS pipeline. HO-Gaussian eliminates the dependency on SfM point initialization, allowing for rendering of urban scenes, and incorporates the Point Densitification to enhance rendering quality in problematic regions during training. Furthermore, we introduce Gaussian Direction Encoding as an alternative for spherical harmonics in the rendering pipeline, which enables view-dependent color representation. To account for multi-camera systems, we introduce neural warping to enhance object consistency across different cameras. Experimental results on widely used autonomous driving datasets demonstrate that HO-Gaussian achieves photo-realistic rendering in real-time on multi-camera urban datasets.
Authors: Tianchen Deng, Yanbo Wang, Hongle Xie, Hesheng Wang, Jingchuan Wang, Danwei Wang, Weidong Chen
Abstract: In recent years, there have been significant advancements in 3D reconstruction and dense RGB-D SLAM systems. One notable development is the application of Neural Radiance Fields (NeRF) in these systems, which utilizes implicit neural representation to encode 3D scenes. This extension of NeRF to SLAM has shown promising results. However, the depth images obtained from consumer-grade RGB-D sensors are often sparse and noisy, which poses significant challenges for 3D reconstruction and affects the accuracy of the representation of the scene geometry. Moreover, the original hierarchical feature grid with occupancy value is inaccurate for scene geometry representation. Furthermore, the existing methods select random pixels for camera tracking, which leads to inaccurate localization and is not robust in real-world indoor environments. To this end, we present NeSLAM, an advanced framework that achieves accurate and dense depth estimation, robust camera tracking, and realistic synthesis of novel views. First, a depth completion and denoising network is designed to provide dense geometry prior and guide the neural implicit representation optimization. Second, the occupancy scene representation is replaced with Signed Distance Field (SDF) hierarchical scene representation for high-quality reconstruction and view synthesis. Furthermore, we also propose a NeRF-based self-supervised feature tracking algorithm for robust real-time tracking. Experiments on various indoor datasets demonstrate the effectiveness and accuracy of the system in reconstruction, tracking quality, and novel view synthesis.
Authors: Xue Jiang, Feng Liu, Zhen Fang, Hong Chen, Tongliang Liu, Feng Zheng, Bo Han
Abstract: Out-of-distribution (OOD) detection aims at identifying samples from unknown classes, playing a crucial role in trustworthy models against errors on unexpected inputs. Extensive research has been dedicated to exploring OOD detection in the vision modality. Vision-language models (VLMs) can leverage both textual and visual information for various multi-modal applications, whereas few OOD detection methods take into account information from the text modality. In this paper, we propose a novel post hoc OOD detection method, called NegLabel, which takes a vast number of negative labels from extensive corpus databases. We design a novel scheme for the OOD score collaborated with negative labels. Theoretical analysis helps to understand the mechanism of negative labels. Extensive experiments demonstrate that our method NegLabel achieves state-of-the-art performance on various OOD detection benchmarks and generalizes well on multiple VLM architectures. Furthermore, our method NegLabel exhibits remarkable robustness against diverse domain shifts. The codes are available at https://github.com/tmlr-group/NegLabel.
Authors: Zhongrui Yu, Haoran Wang, Jinze Yang, Hanzhang Wang, Zeke Xie, Yunfeng Cai, Jiale Cao, Zhong Ji, Mingming Sun
Abstract: Novel View Synthesis (NVS) for street scenes play a critical role in the autonomous driving simulation. The current mainstream technique to achieve it is neural rendering, such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS). Although thrilling progress has been made, when handling street scenes, current methods struggle to maintain rendering quality at the viewpoint that deviates significantly from the training viewpoints. This issue stems from the sparse training views captured by a fixed camera on a moving vehicle. To tackle this problem, we propose a novel approach that enhances the capacity of 3DGS by leveraging prior from a Diffusion Model along with complementary multi-modal data. Specifically, we first fine-tune a Diffusion Model by adding images from adjacent frames as condition, meanwhile exploiting depth data from LiDAR point clouds to supply additional spatial information. Then we apply the Diffusion Model to regularize the 3DGS at unseen views during training. Experimental results validate the effectiveness of our method compared with current state-of-the-art models, and demonstrate its advance in rendering images from broader views.
Authors: Yuiko Sakuma, Masakazu Yoshimura, Junji Otsuka, Atsushi Irie, Takeshi Ohashi
Abstract: Compression of large and performant vision foundation models (VFMs) into arbitrary bit-wise operations (BitOPs) allows their deployment on various hardware. We propose to fine-tune a VFM to a mixed-precision quantized supernet. The supernet-based neural architecture search (NAS) can be adopted for this purpose, which trains a supernet, and then subnets within arbitrary hardware budgets can be extracted. However, existing methods face difficulties in optimizing the mixed-precision search space and incurring large memory costs during training. To tackle these challenges, first, we study the effective search space design for fine-tuning a VFM by comparing different operators (such as resolution, feature size, width, depth, and bit-widths) in terms of performance and BitOPs reduction. Second, we propose memory-efficient supernet training using a low-rank adapter (LoRA) and a progressive training strategy. The proposed method is evaluated for the recently proposed VFM, Segment Anything Model, fine-tuned on segmentation tasks. The searched model yields about a 95% reduction in BitOPs without incurring performance degradation.
Authors: Giovanni Bellitto, Federica Proietto Salanitri, Matteo Pennisi, Matteo Boschini, Angelo Porrello, Simone Calderara, Simone Palazzo, Concetto Spampinato
Abstract: We present SAM, a biologically-plausible selective attention-driven modulation approach to enhance classification models in a continual learning setting. Inspired by neurophysiological evidence that the primary visual cortex does not contribute to object manifold untangling for categorization and that primordial attention biases are still embedded in the modern brain, we propose to employ auxiliary saliency prediction features as a modulation signal to drive and stabilize the learning of a sequence of non-i.i.d. classification tasks. Experimental results confirm that SAM effectively enhances the performance (in some cases up to about twenty percent points) of state-of-the-art continual learning methods, both in class-incremental and task-incremental settings. Moreover, we show that attention-based modulation successfully encourages the learning of features that are more robust to the presence of spurious features and to adversarial attacks than baseline methods. Code is available at: https://github.com/perceivelab/SAM.
Authors: Qi Bi, Shaodi You, Theo Gevers
Abstract: Images from outdoor scenes may be taken under various weather conditions. It is well studied that weather impacts the performance of computer vision algorithms and needs to be handled properly. However, existing algorithms model weather condition as a discrete status and estimate it using multi-label classification. The fact is that, physically, specifically in meteorology, weather are modeled as a continuous and transitional status. Instead of directly implementing hard classification as existing multi-weather classification methods do, we consider the physical formulation of multi-weather conditions and model the impact of physical-related parameter on learning from the image appearance. In this paper, we start with solid revisit of the physics definition of weather and how it can be described as a continuous machine learning and computer vision task. Namely, we propose to model the weather uncertainty, where the level of probability and co-existence of multiple weather conditions are both considered. A Gaussian mixture model is used to encapsulate the weather uncertainty and a uncertainty-aware multi-weather learning scheme is proposed based on prior-posterior learning. A novel multi-weather co-presence estimation transformer (MeFormer) is proposed. In addition, a new multi-weather co-presence estimation (MePe) dataset, along with 14 fine-grained weather categories and 16,078 samples, is proposed to benchmark both conventional multi-label weather classification task and multi-weather co-presence estimation task. Large scale experiments show that the proposed method achieves state-of-the-art performance and substantial generalization capabilities on both the conventional multi-label weather classification task and the proposed multi-weather co-presence estimation task. Besides, modeling weather uncertainty also benefits adverse-weather semantic segmentation.
Authors: Barbara Toniella Corradini, Mustafa Shukor, Paul Couairon, Guillaume Couairon, Franco Scarselli, Matthieu Cord
Abstract: Foundation models have exhibited unprecedented capabilities in tackling many domains and tasks. Models such as CLIP are currently widely used to bridge cross-modal representations, and text-to-image diffusion models are arguably the leading models in terms of realistic image generation. Image generative models are trained on massive datasets that provide them with powerful internal spatial representations. In this work, we explore the potential benefits of such representations, beyond image generation, in particular, for dense visual prediction tasks. We focus on the task of image segmentation, which is traditionally solved by training models on closed-vocabulary datasets, with pixel-level annotations. To avoid the annotation cost or training large diffusion models, we constraint our setup to be zero-shot and training-free. In a nutshell, our pipeline leverages different and relatively small-sized, open-source foundation models for zero-shot open-vocabulary segmentation. The pipeline is as follows: the image is passed to both a captioner model (i.e. BLIP) and a diffusion model (i.e., Stable Diffusion Model) to generate a text description and visual representation, respectively. The features are clustered and binarized to obtain class agnostic masks for each object. These masks are then mapped to a textual class, using the CLIP model to support open-vocabulary. Finally, we add a refinement step that allows to obtain a more precise segmentation mask. Our approach (dubbed FreeSeg-Diff), which does not rely on any training, outperforms many training-based approaches on both Pascal VOC and COCO datasets. In addition, we show very competitive results compared to the recent weakly-supervised segmentation approaches. We provide comprehensive experiments showing the superiority of diffusion model features compared to other pretrained models. Project page: https://bcorrad.github.io/freesegdiff/
Authors: Hu Gao, Depeng Dang
Abstract: Image deblurring is a process of restoring a high quality image from the corresponding blurred image. Significant progress in this field has been made possible by the emergence of various effective deep learning models, including CNNs and Transformers. However, these methods often face the dilemma between eliminating long-range blur degradation perturbations and maintaining computational efficiency, which hinders their practical application. To address this issue, we propose an efficient image deblurring network that leverages selective structured state spaces model to aggregate enriched and accurate features. Specifically, we design an aggregate local and global block (ALGBlock) to capture and fuse both local invariant properties and non-local information. The ALGBlock consists of two blocks: (1) The local block models local connectivity using simplified channel attention. (2) The global block captures long-range dependency features with linear complexity through selective structured state spaces. Nevertheless, we note that the image details are local features of images, we accentuate the local part for restoration by recalibrating the weight when aggregating the two branches for recovery. Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches on widely used benchmarks, highlighting its superior performance.
Authors: David V\'azquez-Lema (University of Coru\~na), Eduardo Mosqueira-Rey (University of Coru\~na), Elena Hern\'andez-Pereira (University of Coru\~na), Carlos Fern\'andez-Lozano (University of Coru\~na), Fernando Seara-Romera (University of Coru\~na), Jorge Pombo-Otero (Complejo Hospitalario Universitario de A Coru\~na)
Abstract: This paper explores the application of Human-in-the-Loop (HITL) strategies in training machine learning models in the medical domain. In this case a doctor-in-the-loop approach is proposed to leverage human expertise in dealing with large and complex data. Specifically, the paper deals with the integration of genomic data and Whole Slide Imaging (WSI) analysis of breast cancer. Three different tasks were developed: segmentation of histopathological images, classification of this images regarding the genomic subtype of the cancer and, finally, interpretation of the machine learning results. The involvement of a pathologist helped us to develop a better segmentation model and to enhance the explainatory capabilities of the models, but the classification results were suboptimal, highlighting the limitations of this approach: despite involving human experts, complex domains can still pose challenges, and a HITL approach may not always be effective.
Authors: Beomyoung Kim, Joonsang Yu, Sung Ju Hwang
Abstract: Panoptic segmentation, combining semantic and instance segmentation, stands as a cutting-edge computer vision task. Despite recent progress with deep learning models, the dynamic nature of real-world applications necessitates continual learning, where models adapt to new classes (plasticity) over time without forgetting old ones (catastrophic forgetting). Current continual segmentation methods often rely on distillation strategies like knowledge distillation and pseudo-labeling, which are effective but result in increased training complexity and computational overhead. In this paper, we introduce a novel and efficient method for continual panoptic segmentation based on Visual Prompt Tuning, dubbed ECLIPSE. Our approach involves freezing the base model parameters and fine-tuning only a small set of prompt embeddings, addressing both catastrophic forgetting and plasticity and significantly reducing the trainable parameters. To mitigate inherent challenges such as error propagation and semantic drift in continual segmentation, we propose logit manipulation to effectively leverage common knowledge across the classes. Experiments on ADE20K continual panoptic segmentation benchmark demonstrate the superiority of ECLIPSE, notably its robustness against catastrophic forgetting and its reasonable plasticity, achieving a new state-of-the-art. The code is available at https://github.com/clovaai/ECLIPSE.
Authors: Sidi Wu, Yizi Chen, Samuel Mermet, Lorenz Hurni, Konrad Schindler, Nicolas Gonthier, Loic Landrieu
Abstract: Most image-to-image translation models postulate that a unique correspondence exists between the semantic classes of the source and target domains. However, this assumption does not always hold in real-world scenarios due to divergent distributions, different class sets, and asymmetrical information representation. As conventional GANs attempt to generate images that match the distribution of the target domain, they may hallucinate spurious instances of classes absent from the source domain, thereby diminishing the usefulness and reliability of translated images. CycleGAN-based methods are also known to hide the mismatched information in the generated images to bypass cycle consistency objectives, a process known as steganography. In response to the challenge of non-bijective image translation, we introduce StegoGAN, a novel model that leverages steganography to prevent spurious features in generated images. Our approach enhances the semantic consistency of the translated images without requiring additional postprocessing or supervision. Our experimental evaluations demonstrate that StegoGAN outperforms existing GAN-based models across various non-bijective image-to-image translation tasks, both qualitatively and quantitatively. Our code and pretrained models are accessible at https://github.com/sian-wusidi/StegoGAN.
Authors: Jaehoon Ko, Kyusun Cho, Joungbin Lee, Heeji Yoon, Sangmin Lee, Sangjun Ahn, Seungryong Kim
Abstract: Recent methods for audio-driven talking head synthesis often optimize neural radiance fields (NeRF) on a monocular talking portrait video, leveraging its capability to render high-fidelity and 3D-consistent novel-view frames. However, they often struggle to reconstruct complete face geometry due to the absence of comprehensive 3D information in the input monocular videos. In this paper, we introduce a novel audio-driven talking head synthesis framework, called Talk3D, that can faithfully reconstruct its plausible facial geometries by effectively adopting the pre-trained 3D-aware generative prior. Given the personalized 3D generative model, we present a novel audio-guided attention U-Net architecture that predicts the dynamic face variations in the NeRF space driven by audio. Furthermore, our model is further modulated by audio-unrelated conditioning tokens which effectively disentangle variations unrelated to audio features. Compared to existing methods, our method excels in generating realistic facial geometries even under extreme head poses. We also conduct extensive experiments showing our approach surpasses state-of-the-art benchmarks in terms of both quantitative and qualitative evaluations.
Authors: Ke Wu, Kaizhao Zhang, Zhiwei Zhang, Shanshuai Yuan, Muer Tie, Julong Wei, Zijun Xu, Jieru Zhao, Zhongxue Gan, Wenchao Ding
Abstract: Online dense mapping of urban scenes forms a fundamental cornerstone for scene understanding and navigation of autonomous vehicles. Recent advancements in mapping methods are mainly based on NeRF, whose rendering speed is too slow to meet online requirements. 3D Gaussian Splatting (3DGS), with its rendering speed hundreds of times faster than NeRF, holds greater potential in online dense mapping. However, integrating 3DGS into a street-view dense mapping framework still faces two challenges, including incomplete reconstruction due to the absence of geometric information beyond the LiDAR coverage area and extensive computation for reconstruction in large urban scenes. To this end, we propose HGS-Mapping, an online dense mapping framework in unbounded large-scale scenes. To attain complete construction, our framework introduces Hybrid Gaussian Representation, which models different parts of the entire scene using Gaussians with distinct properties. Furthermore, we employ a hybrid Gaussian initialization mechanism and an adaptive update method to achieve high-fidelity and rapid reconstruction. To the best of our knowledge, we are the first to integrate Gaussian representation into online dense mapping of urban scenes. Our approach achieves SOTA reconstruction accuracy while only employing 66% number of Gaussians, leading to 20% faster reconstruction speed.
Authors: Qiang Guo, Rubo Zhang, Di Zhao
Abstract: Aiming at the metro video surveillance system has not been able to effectively solve the metro crowd density estimation problem, a Metro Crowd density estimation Network (called MCNet) is proposed to automatically classify crowd density level of passengers. Firstly, an Integrating Multi-scale Attention (IMA) module is proposed to enhance the ability of the plain classifiers to extract semantic crowd texture features to accommodate to the characteristics of the crowd texture feature. The innovation of the IMA module is to fuse the dilation convolution, multiscale feature extraction and attention mechanism to obtain multi-scale crowd feature activation from a larger receptive field with lower computational cost, and to strengthen the crowds activation state of convolutional features in top layers. Secondly, a novel lightweight crowd texture feature extraction network is proposed, which can directly process video frames and automatically extract texture features for crowd density estimation, while its faster image processing speed and fewer network parameters make it flexible to be deployed on embedded platforms with limited hardware resources. Finally, this paper integrates IMA module and the lightweight crowd texture feature extraction network to construct the MCNet, and validate the feasibility of this network on image classification dataset: Cifar10 and four crowd density datasets: PETS2009, Mall, QUT and SH_METRO to validate the MCNet whether can be a suitable solution for crowd density estimation in metro video surveillance where there are image processing challenges such as high density, high occlusion, perspective distortion and limited hardware resources.
Authors: Shuangjian Li, Tao Zhu, Furong Duan, Liming Chen, Huansheng Ning, Yaping Wan
Abstract: Wearable sensor human activity recognition (HAR) is a crucial area of research in activity sensing. While transformer-based temporal deep learning models have been extensively studied and implemented, their large number of parameters present significant challenges in terms of system computing load and memory usage, rendering them unsuitable for real-time mobile activity recognition applications. Recently, an efficient hardware-aware state space model (SSM) called Mamba has emerged as a promising alternative. Mamba demonstrates strong potential in long sequence modeling, boasts a simpler network architecture, and offers an efficient hardware-aware design. Leveraging SSM for activity recognition represents an appealing avenue for exploration. In this study, we introduce HARMamba, which employs a more lightweight selective SSM as the foundational model architecture for activity recognition. The goal is to address the computational resource constraints encountered in real-time activity recognition scenarios. Our approach involves processing sensor data flow by independently learning each channel and segmenting the data into "patches". The marked sensor sequence's position embedding serves as the input token for the bidirectional state space model, ultimately leading to activity categorization through the classification head. Compared to established activity recognition frameworks like Transformer-based models, HARMamba achieves superior performance while also reducing computational and memory overhead. Furthermore, our proposed method has been extensively tested on four public activity datasets: PAMAP2, WISDM, UNIMIB, and UCI, demonstrating impressive performance in activity recognition tasks.
Authors: Pengzhi Li, Baijuan Li, Zhiheng Li
Abstract: Recently, the development of large-scale models has paved the way for various interdisciplinary research, including architecture. By using generative AI, we present a novel workflow that utilizes AI models to generate conceptual floorplans and 3D models from simple sketches, enabling rapid ideation and controlled generation of architectural renderings based on textual descriptions. Our work demonstrates the potential of generative AI in the architectural design process, pointing towards a new direction of computer-aided architectural design. Our project website is available at: https://zrealli.github.io/sketch2arc
Authors: Luozhou Wang, Guibao Shen, Yixun Liang, Xin Tao, Pengfei Wan, Di Zhang, Yijun Li, Yingcong Chen
Abstract: In this research, we present a novel approach to motion customization in video generation, addressing the widespread gap in the thorough exploration of motion representation within video generative models. Recognizing the unique challenges posed by video's spatiotemporal nature, our method introduces Motion Embeddings, a set of explicit, temporally coherent one-dimensional embeddings derived from a given video. These embeddings are designed to integrate seamlessly with the temporal transformer modules of video diffusion models, modulating self-attention computations across frames without compromising spatial integrity. Our approach offers a compact and efficient solution to motion representation and enables complex manipulations of motion characteristics through vector arithmetic in the embedding space. Furthermore, we identify the Temporal Discrepancy in video generative models, which refers to variations in how different motion modules process temporal relationships between frames. We leverage this understanding to optimize the integration of our motion embeddings. Our contributions include the introduction of a tailored motion embedding for customization tasks, insights into the temporal processing differences in video models, and a demonstration of the practical advantages and effectiveness of our method through extensive experiments.
Authors: Victor Silva dos Santos, Erwan Gloaguen, Shiva Tirdad
Abstract: Geological maps are an extremely valuable source of information for the Earth sciences. They provide insights into mineral exploration, vulnerability to natural hazards, and many other applications. These maps are created using numerical or conceptual models that use geological observations to extrapolate data. Geostatistical techniques have traditionally been used to generate reliable predictions that take into account the spatial patterns inherent in the data. However, as the number of auxiliary variables increases, these methods become more labor-intensive. Additionally, traditional machine learning methods often struggle with spatially correlated data and extracting valuable non-linear information from geoscientific datasets. To address these limitations, a new architecture called the Spatially Constrained Bayesian Network (SCB-Net) has been developed. The SCB-Net aims to effectively exploit the information from auxiliary variables while producing spatially constrained predictions. It is made up of two parts, the first part focuses on learning underlying patterns in the auxiliary variables while the second part integrates ground-truth data and the learned embeddings from the first part. Moreover, to assess model uncertainty, a technique called Monte Carlo dropout is used as a Bayesian approximation. The SCB-Net has been applied to two selected areas in northern Quebec, Canada, and has demonstrated its potential in generating field-data-constrained lithological maps while allowing assessment of prediction uncertainty for decision-making. This study highlights the promising advancements of deep neural networks in geostatistics, particularly in handling complex spatial feature learning tasks, leading to improved spatial information techniques.
Authors: Chao Pang, Jiang Wu, Jiayu Li, Yi Liu, Jiaxing Sun, Weijia Li, Xingxing Weng, Shuai Wang, Litong Feng, Gui-Song Xia, Conghui He
Abstract: The generic large Vision-Language Models (VLMs) is rapidly developing, but still perform poorly in Remote Sensing (RS) domain, which is due to the unique and specialized nature of RS imagery and the comparatively limited spatial perception of current VLMs. Existing Remote Sensing specific Vision Language Models (RSVLMs) still have considerable potential for improvement, primarily owing to the lack of large-scale, high-quality RS vision-language datasets. We constructed HqDC-1.4M, the large scale High quality and Detailed Captions for RS images, containing 1.4 million image-caption pairs, which not only enhance the RSVLM's understanding of RS images but also significantly improve the model's spatial perception abilities, such as localization and counting, thereby increasing the helpfulness of the RSVLM. Moreover, to address the inevitable "hallucination" problem in RSVLM, we developed RSSA, the first dataset aimed at enhancing the Self-Awareness capability of RSVLMs. By incorporating a variety of unanswerable questions into typical RS visual question-answering tasks, RSSA effectively improves the truthfulness and reduces the hallucinations of the model's outputs, thereby enhancing the honesty of the RSVLM. Based on these datasets, we proposed the H2RSVLM, the Helpful and Honest Remote Sensing Vision Language Model. H2RSVLM has achieved outstanding performance on multiple RS public datasets and is capable of recognizing and refusing to answer the unanswerable questions, effectively mitigating the incorrect generations. We will release the code, data and model weights at https://github.com/opendatalab/H2RSVLM .
Authors: Sanghyun Woo, Kwanyong Park, Inkyu Shin, Myungchul Kim, In So Kweon
Abstract: Multi-target multi-camera tracking is a crucial task that involves identifying and tracking individuals over time using video streams from multiple cameras. This task has practical applications in various fields, such as visual surveillance, crowd behavior analysis, and anomaly detection. However, due to the difficulty and cost of collecting and labeling data, existing datasets for this task are either synthetically generated or artificially constructed within a controlled camera network setting, which limits their ability to model real-world dynamics and generalize to diverse camera configurations. To address this issue, we present MTMMC, a real-world, large-scale dataset that includes long video sequences captured by 16 multi-modal cameras in two different environments - campus and factory - across various time, weather, and season conditions. This dataset provides a challenging test-bed for studying multi-camera tracking under diverse real-world complexities and includes an additional input modality of spatially aligned and temporally synchronized RGB and thermal cameras, which enhances the accuracy of multi-camera tracking. MTMMC is a super-set of existing datasets, benefiting independent fields such as person detection, re-identification, and multiple object tracking. We provide baselines and new learning setups on this dataset and set the reference scores for future studies. The datasets, models, and test server will be made publicly available.
Authors: You Wu, Kean Liu, Xiaoyue Mi, Fan Tang, Juan Cao, Jintao Li
Abstract: Concept personalization methods enable large text-to-image models to learn specific subjects (e.g., objects/poses/3D models) and synthesize renditions in new contexts. Given that the image references are highly biased towards visual attributes, state-of-the-art personalization models tend to overfit the whole subject and cannot disentangle visual characteristics in pixel space. In this study, we proposed a more challenging setting, namely fine-grained visual appearance personalization. Different from existing methods, we allow users to provide a sentence describing the desired attributes. A novel decoupled self-augmentation strategy is proposed to generate target-related and non-target samples to learn user-specified visual attributes. These augmented data allow for refining the model's understanding of the target attribute while mitigating the impact of unrelated attributes. At the inference stage, adjustments are conducted on semantic space through the learned target and non-target embeddings to further enhance the disentanglement of target attributes. Extensive experiments on various kinds of visual attributes with SOTA personalization methods show the ability of the proposed method to mimic target visual appearance in novel contexts, thus improving the controllability and flexibility of personalization.
Authors: Chih-Hui Ho, Kuan-Chuan Peng, Nuno Vasconcelos
Abstract: Anomaly detection (AD) aims to identify defective images and localize their defects (if any). Ideally, AD models should be able to detect defects over many image classes; without relying on hard-coded class names that can be uninformative or inconsistent across datasets; learn without anomaly supervision; and be robust to the long-tailed distributions of real-world applications. To address these challenges, we formulate the problem of long-tailed AD by introducing several datasets with different levels of class imbalance and metrics for performance evaluation. We then propose a novel method, LTAD, to detect defects from multiple and long-tailed classes, without relying on dataset class names. LTAD combines AD by reconstruction and semantic AD modules. AD by reconstruction is implemented with a transformer-based reconstruction module. Semantic AD is implemented with a binary classifier, which relies on learned pseudo class names and a pretrained foundation model. These modules are learned over two phases. Phase 1 learns the pseudo-class names and a variational autoencoder (VAE) for feature synthesis that augments the training data to combat long-tails. Phase 2 then learns the parameters of the reconstruction and classification modules of LTAD. Extensive experiments using the proposed long-tailed datasets show that LTAD substantially outperforms the state-of-the-art methods for most forms of dataset imbalance. The long-tailed dataset split is available at https://zenodo.org/records/10854201 .
Authors: Yinwei Wu, Xingyi Yang, Xinchao Wang
Abstract: Despite their exceptional generative abilities, large text-to-image diffusion models, much like skilled but careless artists, often struggle with accurately depicting visual relationships between objects. This issue, as we uncover through careful analysis, arises from a misaligned text encoder that struggles to interpret specific relationships and differentiate the logical order of associated objects. To resolve this, we introduce a novel task termed Relation Rectification, aiming to refine the model to accurately represent a given relationship it initially fails to generate. To address this, we propose an innovative solution utilizing a Heterogeneous Graph Convolutional Network (HGCN). It models the directional relationships between relation terms and corresponding objects within the input prompts. Specifically, we optimize the HGCN on a pair of prompts with identical relational words but reversed object orders, supplemented by a few reference images. The lightweight HGCN adjusts the text embeddings generated by the text encoder, ensuring the accurate reflection of the textual relation in the embedding space. Crucially, our method retains the parameters of the text encoder and diffusion model, preserving the model's robust performance on unrelated descriptions. We validated our approach on a newly curated dataset of diverse relational data, demonstrating both quantitative and qualitative enhancements in generating images with precise visual relations. Project page: https://wuyinwei-hah.github.io/rrnet.github.io/.
Authors: Jos\'e Celestino, Manuel Marques, Jacinto C. Nascimento
Abstract: Head pose estimation has become a crucial area of research in computer vision given its usefulness in a wide range of applications, including robotics, surveillance, or driver attention monitoring. One of the most difficult challenges in this field is managing head occlusions that frequently take place in real-world scenarios. In this paper, we propose a novel and efficient framework that is robust in real world head occlusion scenarios. In particular, we propose an unsupervised latent embedding clustering with regression and classification components for each pose angle. The model optimizes latent feature representations for occluded and non-occluded images through a clustering term while improving fine-grained angle predictions. Experimental evaluation on in-the-wild head pose benchmark datasets reveal competitive performance in comparison to state-of-the-art methodologies with the advantage of having a significant data reduction. We observe a substantial improvement in occluded head pose estimation. Also, an ablation study is conducted to ascertain the impact of the clustering term within our proposed framework.
Authors: Taha Koleilat, Hojat Asgariandehkordi, Hassan Rivaz, Yiming Xiao
Abstract: Medical image segmentation of anatomical structures and pathology is crucial in modern clinical diagnosis, disease study, and treatment planning. To date, great progress has been made in deep learning-based segmentation techniques, but most methods still lack data efficiency, generalizability, and interactability. Consequently, the development of new, precise segmentation methods that demand fewer labeled datasets is of utmost importance in medical image analysis. Recently, the emergence of foundation models, such as CLIP and Segment-Anything-Model (SAM), with comprehensive cross-domain representation opened the door for interactive and universal image segmentation. However, exploration of these models for data-efficient medical image segmentation is still limited, but is highly necessary. In this paper, we propose a novel framework, called MedCLIP-SAM that combines CLIP and SAM models to generate segmentation of clinical scans using text prompts in both zero-shot and weakly supervised settings. To achieve this, we employed a new Decoupled Hard Negative Noise Contrastive Estimation (DHN-NCE) loss to fine-tune the BiomedCLIP model and the recent gScoreCAM to generate prompts to obtain segmentation masks from SAM in a zero-shot setting. Additionally, we explored the use of zero-shot segmentation labels in a weakly supervised paradigm to improve the segmentation quality further. By extensively testing three diverse segmentation tasks and medical image modalities (breast tumor ultrasound, brain tumor MRI, and lung X-ray), our proposed framework has demonstrated excellent accuracy.
Authors: Runhao Zeng, Xiaoyong Chen, Jiaming Liang, Huisi Wu, Guangzhong Cao, Yong Guo
Abstract: Temporal action detection (TAD) aims to locate action positions and recognize action categories in long-term untrimmed videos. Although many methods have achieved promising results, their robustness has not been thoroughly studied. In practice, we observe that temporal information in videos can be occasionally corrupted, such as missing or blurred frames. Interestingly, existing methods often incur a significant performance drop even if only one frame is affected. To formally evaluate the robustness, we establish two temporal corruption robustness benchmarks, namely THUMOS14-C and ActivityNet-v1.3-C. In this paper, we extensively analyze the robustness of seven leading TAD methods and obtain some interesting findings: 1) Existing methods are particularly vulnerable to temporal corruptions, and end-to-end methods are often more susceptible than those with a pre-trained feature extractor; 2) Vulnerability mainly comes from localization error rather than classification error; 3) When corruptions occur in the middle of an action instance, TAD models tend to yield the largest performance drop. Besides building a benchmark, we further develop a simple but effective robust training method to defend against temporal corruptions, through the FrameDrop augmentation and Temporal-Robust Consistency loss. Remarkably, our approach not only improves robustness but also yields promising improvements on clean data. We believe that this study will serve as a benchmark for future research in robust video analysis. Source code and models are available at https://github.com/Alvin-Zeng/temporal-robustness-benchmark.
URLs: https://github.com/Alvin-Zeng/temporal-robustness-benchmark.
Authors: Shreyasi Pathak, J\"org Schl\"otterer, Jeroen Veltman, Jeroen Geerdink, Maurice van Keulen, Christin Seifert
Abstract: Deep learning models have achieved high performance in medical applications, however, their adoption in clinical practice is hindered due to their black-box nature. Self-explainable models, like prototype-based models, can be especially beneficial as they are interpretable by design. However, if the learnt prototypes are of low quality then the prototype-based models are as good as black-box. Having high quality prototypes is a pre-requisite for a truly interpretable model. In this work, we propose a prototype evaluation framework for coherence (PEF-C) for quantitatively evaluating the quality of the prototypes based on domain knowledge. We show the use of PEF-C in the context of breast cancer prediction using mammography. Existing works on prototype-based models on breast cancer prediction using mammography have focused on improving the classification performance of prototype-based models compared to black-box models and have evaluated prototype quality through anecdotal evidence. We are the first to go beyond anecdotal evidence and evaluate the quality of the mammography prototypes systematically using our PEF-C. Specifically, we apply three state-of-the-art prototype-based models, ProtoPNet, BRAIxProtoPNet++ and PIP-Net on mammography images for breast cancer prediction and evaluate these models w.r.t. i) classification performance, and ii) quality of the prototypes, on three public datasets. Our results show that prototype-based models are competitive with black-box models in terms of classification performance, and achieve a higher score in detecting ROIs. However, the quality of the prototypes are not yet sufficient and can be improved in aspects of relevance, purity and learning a variety of prototypes. We call the XAI community to systematically evaluate the quality of the prototypes to check their true usability in high stake decisions and improve such models further.
Authors: Weifeng Lin, Xinyu Wei, Ruichuan An, Peng Gao, Bocheng Zou, Yulin Luo, Siyuan Huang, Shanghang Zhang, Hongsheng Li
Abstract: The interaction between humans and artificial intelligence (AI) is a crucial factor that reflects the effectiveness of multimodal large language models (MLLMs). However, current MLLMs primarily focus on image-level comprehension and limit interaction to textual instructions, thereby constraining their flexibility in usage and depth of response. In this paper, we introduce the Draw-and-Understand project: a new model, a multi-domain dataset, and a challenging benchmark for visual prompting. Specifically, we propose SPHINX-V, a new end-to-end trained Multimodal Large Language Model (MLLM) that connects a vision encoder, a visual prompt encoder and an LLM for various visual prompts (points, bounding boxes, and free-form shape) and language understanding. To advance visual prompting research for MLLMs, we introduce MDVP-Data and MDVP-Bench. MDVP-Data features a multi-domain dataset containing 1.6M unique image-visual prompt-text instruction-following samples, including natural images, document images, OCR images, mobile screenshots, web screenshots, and multi-panel images. Furthermore, we present MDVP-Bench, a comprehensive and challenging benchmark to assess a model's capability in understanding visual prompting instructions. Our experiments demonstrate SPHINX-V's impressive multimodal interaction capabilities through visual prompting, revealing significant improvements in detailed pixel-level description and question-answering abilities.
Authors: Wenyu Yang, Sergio Vitale, Hossein Aghababaei, Giampaolo Ferraioli, Vito Pascazio, Gilda Schirinzi
Abstract: Tropical forests are a key component of the global carbon cycle. With plans for upcoming space-borne missions like BIOMASS to monitor forestry, several airborne missions, including TropiSAR and AfriSAR campaigns, have been successfully launched and experimented. Typical Synthetic Aperture Radar Tomography (TomoSAR) methods involve complex models with low accuracy and high computation costs. In recent years, deep learning methods have also gained attention in the TomoSAR framework, showing interesting performance. Recently, a solution based on a fully connected Tomographic Neural Network (TSNN) has demonstrated its effectiveness in accurately estimating forest and ground heights by exploiting the pixel-wise elements of the covariance matrix derived from TomoSAR data. This work instead goes beyond the pixel-wise approach to define a context-aware deep learning-based solution named CATSNet. A convolutional neural network is considered to leverage patch-based information and extract features from a neighborhood rather than focus on a single pixel. The training is conducted by considering TomoSAR data as the input and Light Detection and Ranging (LiDAR) values as the ground truth. The experimental results show striking advantages in both performance and generalization ability by leveraging context information within Multiple Baselines (MB) TomoSAR data across different polarimetric modalities, surpassing existing techniques.
Authors: Mauro Comi, Alessio Tonioni, Max Yang, Jonathan Tremblay, Valts Blukis, Yijiong Lin, Nathan F. Lepora, Laurence Aitchison
Abstract: Touch and vision go hand in hand, mutually enhancing our ability to understand the world. From a research perspective, the problem of mixing touch and vision is underexplored and presents interesting challenges. To this end, we propose Tactile-Informed 3DGS, a novel approach that incorporates touch data (local depth maps) with multi-view vision data to achieve surface reconstruction and novel view synthesis. Our method optimises 3D Gaussian primitives to accurately model the object's geometry at points of contact. By creating a framework that decreases the transmittance at touch locations, we achieve a refined surface reconstruction, ensuring a uniformly smooth depth map. Touch is particularly useful when considering non-Lambertian objects (e.g. shiny or reflective surfaces) since contemporary methods tend to fail to reconstruct with fidelity specular highlights. By combining vision and tactile sensing, we achieve more accurate geometry reconstructions with fewer images than prior methods. We conduct evaluation on objects with glossy and reflective surfaces and demonstrate the effectiveness of our approach, offering significant improvements in reconstruction quality.
Authors: Thomas Melistas, Nikos Spyrou, Nefeli Gkouti, Pedro Sanchez, Athanasios Vlontzos, Giorgos Papanastasiou, Sotirios A. Tsaftaris
Abstract: Counterfactual image generation is pivotal for understanding the causal relations of variables, with applications in interpretability and generation of unbiased synthetic data. However, evaluating image generation is a long-standing challenge in itself. The need to evaluate counterfactual generation compounds on this challenge, precisely because counterfactuals, by definition, are hypothetical scenarios without observable ground truths. In this paper, we present a novel comprehensive framework aimed at benchmarking counterfactual image generation methods. We incorporate metrics that focus on evaluating diverse aspects of counterfactuals, such as composition, effectiveness, minimality of interventions, and image realism. We assess the performance of three distinct conditional image generation model types, based on the Structural Causal Model paradigm. Our work is accompanied by a user-friendly Python package which allows to further evaluate and benchmark existing and future counterfactual image generation methods. Our framework is extendable to additional SCM and other causal methods, generative models, and datasets.
Authors: Zhiwen Fan, Wenyan Cong, Kairun Wen, Kevin Wang, Jian Zhang, Xinghao Ding, Danfei Xu, Boris Ivanovic, Marco Pavone, Georgios Pavlakos, Zhangyang Wang, Yue Wang
Abstract: While novel view synthesis (NVS) has made substantial progress in 3D computer vision, it typically requires an initial estimation of camera intrinsics and extrinsics from dense viewpoints. This pre-processing is usually conducted via a Structure-from-Motion (SfM) pipeline, a procedure that can be slow and unreliable, particularly in sparse-view scenarios with insufficient matched features for accurate reconstruction. In this work, we integrate the strengths of point-based representations (e.g., 3D Gaussian Splatting, 3D-GS) with end-to-end dense stereo models (DUSt3R) to tackle the complex yet unresolved issues in NVS under unconstrained settings, which encompasses pose-free and sparse view challenges. Our framework, InstantSplat, unifies dense stereo priors with 3D-GS to build 3D Gaussians of large-scale scenes from sparseview & pose-free images in less than 1 minute. Specifically, InstantSplat comprises a Coarse Geometric Initialization (CGI) module that swiftly establishes a preliminary scene structure and camera parameters across all training views, utilizing globally-aligned 3D point maps derived from a pre-trained dense stereo pipeline. This is followed by the Fast 3D-Gaussian Optimization (F-3DGO) module, which jointly optimizes the 3D Gaussian attributes and the initialized poses with pose regularization. Experiments conducted on the large-scale outdoor Tanks & Temples datasets demonstrate that InstantSplat significantly improves SSIM (by 32%) while concurrently reducing Absolute Trajectory Error (ATE) by 80%. These establish InstantSplat as a viable solution for scenarios involving posefree and sparse-view conditions. Project page: instantsplat.github.io.
Authors: Jaisidh Singh, Ishaan Shrivastava, Mayank Vatsa, Richa Singh, Aparna Bharati
Abstract: Existing vision-language models (VLMs) treat text descriptions as a unit, confusing individual concepts in a prompt and impairing visual semantic matching and reasoning. An important aspect of reasoning in logic and language is negations. This paper highlights the limitations of popular VLMs such as CLIP, at understanding the implications of negations, i.e., the effect of the word "not" in a given prompt. To enable evaluation of VLMs on fluent prompts with negations, we present CC-Neg, a dataset containing 228,246 images, true captions and their corresponding negated captions. Using CC-Neg along with modifications to the contrastive loss of CLIP, our proposed CoN-CLIP framework, has an improved understanding of negations. This training paradigm improves CoN-CLIP's ability to encode semantics reliably, resulting in 3.85% average gain in top-1 accuracy for zero-shot image classification across 8 datasets. Further, CoN-CLIP outperforms CLIP on challenging compositionality benchmarks such as SugarCREPE by 4.4%, showcasing emergent compositional understanding of objects, relations, and attributes in text. Overall, our work addresses a crucial limitation of VLMs by introducing a dataset and framework that strengthens semantic associations between images and text, demonstrating improved large-scale foundation models with significantly reduced computational cost, promoting efficiency and accessibility.
Authors: Anurag Roy, Riddhiman Moulick, Vinay K. Verma, Saptarshi Ghosh, Abir Das
Abstract: Continual Learning (CL) enables machine learning models to learn from continuously shifting new training data in absence of data from old tasks. Recently, pretrained vision transformers combined with prompt tuning have shown promise for overcoming catastrophic forgetting in CL. These approaches rely on a pool of learnable prompts which can be inefficient in sharing knowledge across tasks leading to inferior performance. In addition, the lack of fine-grained layer specific prompts does not allow these to fully express the strength of the prompts for CL. We address these limitations by proposing ConvPrompt, a novel convolutional prompt creation mechanism that maintains layer-wise shared embeddings, enabling both layer-specific learning and better concept transfer across tasks. The intelligent use of convolution enables us to maintain a low parameter overhead without compromising performance. We further leverage Large Language Models to generate fine-grained text descriptions of each category which are used to get task similarity and dynamically decide the number of prompts to be learned. Extensive experiments demonstrate the superiority of ConvPrompt and improves SOTA by ~3% with significantly less parameter overhead. We also perform strong ablation over various modules to disentangle the importance of different components.
Authors: Abhinav Kumar, Yuliang Guo, Xinyu Huang, Liu Ren, Xiaoming Liu
Abstract: Monocular 3D detectors achieve remarkable performance on cars and smaller objects. However, their performance drops on larger objects, leading to fatal accidents. Some attribute the failures to training data scarcity or their receptive field requirements of large objects. In this paper, we highlight this understudied problem of generalization to large objects. We find that modern frontal detectors struggle to generalize to large objects even on nearly balanced datasets. We argue that the cause of failure is the sensitivity of depth regression losses to noise of larger objects. To bridge this gap, we comprehensively investigate regression and dice losses, examining their robustness under varying error levels and object sizes. We mathematically prove that the dice loss leads to superior noise-robustness and model convergence for large objects compared to regression losses for a simplified case. Leveraging our theoretical insights, we propose SeaBird (Segmentation in Bird's View) as the first step towards generalizing to large objects. SeaBird effectively integrates BEV segmentation on foreground objects for 3D detection, with the segmentation head trained with the dice loss. SeaBird achieves SoTA results on the KITTI-360 leaderboard and improves existing detectors on the nuScenes leaderboard, particularly for large objects. Code and models at https://github.com/abhi1kumar/SeaBird
Authors: Ahmed Agiza, Marina Neseem, Sherief Reda
Abstract: Adapting models pre-trained on large-scale datasets to a variety of downstream tasks is a common strategy in deep learning. Consequently, parameter-efficient fine-tuning methods have emerged as a promising way to adapt pre-trained models to different tasks while training only a minimal number of parameters. While most of these methods are designed for single-task adaptation, parameter-efficient training in Multi-Task Learning (MTL) architectures is still unexplored. In this paper, we introduce MTLoRA, a novel framework for parameter-efficient training of MTL models. MTLoRA employs Task-Agnostic and Task-Specific Low-Rank Adaptation modules, which effectively disentangle the parameter space in MTL fine-tuning, thereby enabling the model to adeptly handle both task specialization and interaction within MTL contexts. We applied MTLoRA to hierarchical-transformer-based MTL architectures, adapting them to multiple downstream dense prediction tasks. Our extensive experiments on the PASCAL dataset show that MTLoRA achieves higher accuracy on downstream tasks compared to fully fine-tuning the MTL model while reducing the number of trainable parameters by 3.6x. Furthermore, MTLoRA establishes a Pareto-optimal trade-off between the number of trainable parameters and the accuracy of the downstream tasks, outperforming current state-of-the-art parameter-efficient training methods in both accuracy and efficiency. Our code is publicly available.
Authors: Lin Chen, Jinsong Li, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Zehui Chen, Haodong Duan, Jiaqi Wang, Yu Qiao, Dahua Lin, Feng Zhao
Abstract: Large vision-language models (LVLMs) have recently achieved rapid progress, sparking numerous studies to evaluate their multi-modal capabilities. However, we dig into current evaluation works and identify two primary issues: 1) Visual content is unnecessary for many samples. The answers can be directly inferred from the questions and options, or the world knowledge embedded in LLMs. This phenomenon is prevalent across current benchmarks. For instance, GeminiPro achieves 42.9% on the MMMU benchmark without any visual input, and outperforms the random choice baseline across six benchmarks over 20% on average. 2) Unintentional data leakage exists in LLM and LVLM training. LLM and LVLM could still answer some visual-necessary questions without visual content, indicating the memorizing of these samples within large-scale training data. For example, Sphinx-X-MoE gets 43.6% on MMMU without accessing images, surpassing its LLM backbone with 17.9%. Both problems lead to misjudgments of actual multi-modal gains and potentially misguide the study of LVLM. To this end, we present MMStar, an elite vision-indispensable multi-modal benchmark comprising 1,500 samples meticulously selected by humans. MMStar benchmarks 6 core capabilities and 18 detailed axes, aiming to evaluate LVLMs' multi-modal capacities with carefully balanced and purified samples. These samples are first roughly selected from current benchmarks with an automated pipeline, human review is then involved to ensure each curated sample exhibits visual dependency, minimal data leakage, and requires advanced multi-modal capabilities. Moreover, two metrics are developed to measure data leakage and actual performance gain in multi-modal training. We evaluate 16 leading LVLMs on MMStar to assess their multi-modal capabilities, and on 7 benchmarks with the proposed metrics to investigate their data leakage and actual multi-modal gain.
Authors: Atsuyuki Miyai, Jingkang Yang, Jingyang Zhang, Yifei Ming, Qing Yu, Go Irie, Yixuan Li, Hai Li, Ziwei Liu, Kiyoharu Aizawa
Abstract: This paper introduces a novel and significant challenge for Vision Language Models (VLMs), termed Unsolvable Problem Detection (UPD). UPD examines the VLM's ability to withhold answers when faced with unsolvable problems in the context of Visual Question Answering (VQA) tasks. UPD encompasses three distinct settings: Absent Answer Detection (AAD), Incompatible Answer Set Detection (IASD), and Incompatible Visual Question Detection (IVQD). To deeply investigate the UPD problem, extensive experiments indicate that most VLMs, including GPT-4V and LLaVA-Next-34B, struggle with our benchmarks to varying extents, highlighting significant room for the improvements. To address UPD, we explore both training-free and training-based solutions, offering new insights into their effectiveness and limitations. We hope our insights, together with future efforts within the proposed UPD settings, will enhance the broader understanding and development of more practical and reliable VLMs.
Authors: Allison Chen, Ilia Sucholutsky, Olga Russakovsky, Thomas L. Griffiths
Abstract: Does language help make sense of the visual world? How important is it to actually see the world rather than having it described with words? These basic questions about the nature of intelligence have been difficult to answer because we only had one example of an intelligent system -- humans -- and limited access to cases that isolated language or vision. However, the development of sophisticated Vision-Language Models (VLMs) by artificial intelligence researchers offers us new opportunities to explore the contributions that language and vision make to learning about the world. We ablate components from the cognitive architecture of these models to identify their contributions to learning new tasks from limited data. We find that a language model leveraging all components recovers a majority of a VLM's performance, despite its lack of visual input, and that language seems to allow this by providing access to prior knowledge and reasoning.
Authors: Jingtao Zhan, Qingyao Ai, Yiqun Liu, Jia Chen, Shaoping Ma
Abstract: Text-to-image generation systems have emerged as revolutionary tools in the realm of artistic creation, offering unprecedented ease in transforming textual prompts into visual art. However, the efficacy of these systems is intricately linked to the quality of user-provided prompts, which often poses a challenge to users unfamiliar with prompt crafting. This paper addresses this challenge by leveraging user reformulation data from interaction logs to develop an automatic prompt reformulation model. Our in-depth analysis of these logs reveals that user prompt reformulation is heavily dependent on the individual user's capability, resulting in significant variance in the quality of reformulation pairs. To effectively use this data for training, we introduce the Capability-aware Prompt Reformulation (CAPR) framework. CAPR innovatively integrates user capability into the reformulation process through two key components: the Conditional Reformulation Model (CRM) and Configurable Capability Features (CCF). CRM reformulates prompts according to a specified user capability, as represented by CCF. The CCF, in turn, offers the flexibility to tune and guide the CRM's behavior. This enables CAPR to effectively learn diverse reformulation strategies across various user capacities and to simulate high-capability user reformulation during inference. Extensive experiments on standard text-to-image generation benchmarks showcase CAPR's superior performance over existing baselines and its remarkable robustness on unseen systems. Furthermore, comprehensive analyses validate the effectiveness of different components. CAPR can facilitate user-friendly interaction with text-to-image systems and make advanced artistic creation more achievable for a broader range of users.
Authors: Ravi Mangal, Nina Narodytska, Divya Gopinath, Boyue Caroline Hu, Anirban Roy, Susmit Jha, Corina Pasareanu
Abstract: Formal analysis of vision-based deep neural networks (DNNs) is highly desirable but it is very challenging due to the difficulty of expressing formal specifications for vision tasks and the lack of efficient verification procedures. In this paper, we propose to leverage emerging multimodal, vision-language, foundation models (VLMs) as a lens through which we can reason about vision models. VLMs have been trained on a large body of images accompanied by their textual description, and are thus implicitly aware of high-level, human-understandable concepts describing the images. We describe a logical specification language $\texttt{Con}_{\texttt{spec}}$ designed to facilitate writing specifications in terms of these concepts. To define and formally check $\texttt{Con}_{\texttt{spec}}$ specifications, we leverage a VLM, which provides a means to encode and efficiently check natural-language properties of vision models. We demonstrate our techniques on a ResNet-based classifier trained on the RIVAL-10 dataset leveraging CLIP as the multimodal model.
Authors: Silpa Vadakkeeveetil Sreelatha, Adarsh Kappiyath, Anjan Dutta
Abstract: When neural networks are trained on biased datasets, they tend to inadvertently learn spurious correlations, leading to challenges in achieving strong generalization and robustness. Current approaches to address such biases typically involve utilizing bias annotations, reweighting based on pseudo-bias labels, or enhancing diversity within bias-conflicting data points through augmentation techniques. We introduce DeNetDM, a novel debiasing method based on the observation that shallow neural networks prioritize learning core attributes, while deeper ones emphasize biases when tasked with acquiring distinct information. Using a training paradigm derived from Product of Experts, we create both biased and debiased branches with deep and shallow architectures and then distill knowledge to produce the target debiased model. Extensive experiments and analyses demonstrate that our approach outperforms current debiasing techniques, achieving a notable improvement of around 5% in three datasets, encompassing both synthetic and real-world data. Remarkably, DeNetDM accomplishes this without requiring annotations pertaining to bias labels or bias types, while still delivering performance on par with supervised counterparts. Furthermore, our approach effectively harnesses the diversity of bias-conflicting points within the data, surpassing previous methods and obviating the need for explicit augmentation-based methods to enhance the diversity of such bias-conflicting points. The source code will be available upon acceptance.
Authors: Pooria Ashrafian, Milad Yazdani, Moein Heidari, Dena Shahriari, Ilker Hacihaliloglu
Abstract: High-quality, large-scale data is essential for robust deep learning models in medical applications, particularly ultrasound image analysis. Diffusion models facilitate high-fidelity medical image generation, reducing the costs associated with acquiring and annotating new images. This paper utilizes recent vision-language models to produce diverse and realistic synthetic echocardiography image data, preserving key features of the original images guided by textual and semantic label maps. Specifically, we investigate three potential avenues: unconditional generation, generation guided by text, and a hybrid approach incorporating both textual and semantic supervision. We show that the rich contextual information present in the synthesized data potentially enhances the accuracy and interpretability of downstream tasks, such as echocardiography segmentation and classification with improved metrics and faster convergence. Our implementation with checkpoints, prompts, and the created synthetic dataset will be publicly available at \href{https://github.com/Pooria90/DiffEcho}{GitHub}.
Authors: Moein Heidari, Reza Azad, Sina Ghorbani Kolahi, Ren\'e Arimond, Leon Niggemeier, Alaa Sulaiman, Afshin Bozorgpour, Ehsan Khodapanah Aghdam, Amirhossein Kazerouni, Ilker Hacihaliloglu, Dorit Merhof
Abstract: Intrigued by the inherent ability of the human visual system to identify salient regions in complex scenes, attention mechanisms have been seamlessly integrated into various Computer Vision (CV) tasks. Building upon this paradigm, Vision Transformer (ViT) networks exploit attention mechanisms for improved efficiency. This review navigates the landscape of redesigned attention mechanisms within ViTs, aiming to enhance their performance. This paper provides a comprehensive exploration of techniques and insights for designing attention mechanisms, systematically reviewing recent literature in the field of CV. This survey begins with an introduction to the theoretical foundations and fundamental concepts underlying attention mechanisms. We then present a systematic taxonomy of various attention mechanisms within ViTs, employing redesigned approaches. A multi-perspective categorization is proposed based on their application, objectives, and the type of attention applied. The analysis includes an exploration of the novelty, strengths, weaknesses, and an in-depth evaluation of the different proposed strategies. This culminates in the development of taxonomies that highlight key properties and contributions. Finally, we gather the reviewed studies along with their available open-source implementations at our \href{https://github.com/mindflow-institue/Awesome-Attention-Mechanism-in-Medical-Imaging}{GitHub}\footnote{\url{https://github.com/xmindflow/Awesome-Attention-Mechanism-in-Medical-Imaging}}. We aim to regularly update it with the most recent relevant papers.
URLs: https://github.com/mindflow-institue/Awesome-Attention-Mechanism-in-Medical-Imaging, https://github.com/xmindflow/Awesome-Attention-Mechanism-in-Medical-Imaging
Authors: Ali Behrouz, Michele Santacatterina, Ramin Zabih
Abstract: Recent advances in deep learning have mainly relied on Transformers due to their data dependency and ability to learn at scale. The attention module in these architectures, however, exhibits quadratic time and space in input size, limiting their scalability for long-sequence modeling. Despite recent attempts to design efficient and effective architecture backbone for multi-dimensional data, such as images and multivariate time series, existing models are either data independent, or fail to allow inter- and intra-dimension communication. Recently, State Space Models (SSMs), and more specifically Selective State Space Models, with efficient hardware-aware implementation, have shown promising potential for long sequence modeling. Motivated by the success of SSMs, we present MambaMixer, a new architecture with data-dependent weights that uses a dual selection mechanism across tokens and channels, called Selective Token and Channel Mixer. MambaMixer connects selective mixers using a weighted averaging mechanism, allowing layers to have direct access to early features. As a proof of concept, we design Vision MambaMixer (ViM2) and Time Series MambaMixer (TSM2) architectures based on the MambaMixer block and explore their performance in various vision and time series forecasting tasks. Our results underline the importance of selective mixing across both tokens and channels. In ImageNet classification, object detection, and semantic segmentation tasks, ViM2 achieves competitive performance with well-established vision models and outperforms SSM-based vision models. In time series forecasting, TSM2 achieves outstanding performance compared to state-of-the-art methods while demonstrating significantly improved computational cost. These results show that while Transformers, cross-channel attention, and MLPs are sufficient for good performance in time series forecasting, neither is necessary.
Authors: David Yevick
Abstract: A simply implemented activation function with even cubic nonlinearity is introduced that increases the accuracy of neural networks without substantial additional computational resources. This is partially enabled through an apparent tradeoff between convergence and accuracy. The activation function generalizes the standard RELU function by introducing additional degrees of freedom through optimizable parameters that enable the degree of nonlinearity to be adjusted. The associated accuracy enhancement is quantified in the context of the MNIST digit data set through a comparison with standard techniques.
Authors: Ardyn Nordstrom, Morgan Nordstrom, Matthew D. Webb
Abstract: This paper details an innovative methodology to integrate image data into traditional econometric models. Motivated by forecasting sales prices for residential real estate, we harness the power of deep learning to add "information" contained in images as covariates. Specifically, images of homes were categorized and encoded using an ensemble of image classifiers (ResNet-50, VGG16, MobileNet, and Inception V3). Unique features presented within each image were further encoded through panoptic segmentation. Forecasts from a neural network trained on the encoded data results in improved out-of-sample predictive power. We also combine these image-based forecasts with standard hedonic real estate property and location characteristics, resulting in a unified dataset. We show that image-based forecasts increase the accuracy of hedonic forecasts when encoded features are regarded as additional covariates. We also attempt to "explain" which covariates the image-based forecasts are most highly correlated with. The study exemplifies the benefits of interdisciplinary methodologies, merging machine learning and econometrics to harness untapped data sources for more accurate forecasting.
Authors: Wanyu Bian, Albert Jang, Fang Liu
Abstract: Using single-task deep learning methods to reconstruct Magnetic Resonance Imaging (MRI) data acquired with different imaging sequences is inherently challenging. The trained deep learning model typically lacks generalizability, and the dissimilarity among image datasets with different types of contrast leads to suboptimal learning performance. This paper proposes a meta-learning approach to efficiently learn image features from multiple MR image datasets. Our algorithm can perform multi-task learning to simultaneously reconstruct MR images acquired using different imaging sequences with different image contrasts. The experiment results demonstrate the ability of our new meta-learning reconstruction method to successfully reconstruct highly-undersampled k-space data from multiple MRI datasets simultaneously, outperforming other compelling reconstruction methods previously developed for single-task learning.
Authors: Hongzhi Liu, Guicheng Li, Jiacheng Nie, Hui Tang, Chunfeng Yang, Qianjin Feng, Hailin Xu, Yang Chen
Abstract: Because of the complicated mechanism of ankle injury, it is very difficult to diagnose ankle fracture in clinic. In order to simplify the process of fracture diagnosis, an automatic diagnosis model of ankle fracture was proposed. Firstly, a tibia-fibula segmentation network is proposed for the joint tibiofibular region of the ankle joint, and the corresponding segmentation dataset is established on the basis of fracture data. Secondly, the image registration method is used to register the bone segmentation mask with the normal bone mask. Finally, a semi-supervised classifier is constructed to make full use of a large number of unlabeled data to classify ankle fractures. Experiments show that the proposed method can segment fractures with fracture lines accurately and has better performance than the general method. At the same time, this method is superior to classification network in several indexes.
Authors: Yunhao Li, Xiaodong Wang, Ping Wang, Xin Yuan, Peidong Liu
Abstract: In this paper, we explore the potential of Snapshot Compressive Imaging (SCI) technique for recovering the underlying 3D scene representation from a single temporal compressed image. SCI is a cost-effective method that enables the recording of high-dimensional data, such as hyperspectral or temporal information, into a single image using low-cost 2D imaging sensors. To achieve this, a series of specially designed 2D masks are usually employed, which not only reduces storage requirements but also offers potential privacy protection. Inspired by this, to take one step further, our approach builds upon the powerful 3D scene representation capabilities of neural radiance fields (NeRF). Specifically, we formulate the physical imaging process of SCI as part of the training of NeRF, allowing us to exploit its impressive performance in capturing complex scene structures. To assess the effectiveness of our method, we conduct extensive evaluations using both synthetic data and real data captured by our SCI system. Extensive experimental results demonstrate that our proposed approach surpasses the state-of-the-art methods in terms of image reconstruction and novel view image synthesis. Moreover, our method also exhibits the ability to restore high frame-rate multi-view consistent images by leveraging SCI and the rendering capabilities of NeRF. The code is available at https://github.com/WU-CVGL/SCINeRF.
Authors: Renkai Wu, Yinghao Liu, Pengchen Liang, Qing Chang
Abstract: Traditionally for improving the segmentation performance of models, most approaches prefer to use adding more complex modules. And this is not suitable for the medical field, especially for mobile medical devices, where computationally loaded models are not suitable for real clinical environments due to computational resource constraints. Recently, state-space models (SSMs), represented by Mamba, have become a strong competitor to traditional CNNs and Transformers. In this paper, we deeply explore the key elements of parameter influence in Mamba and propose an UltraLight Vision Mamba UNet (UltraLight VM-UNet) based on this. Specifically, we propose a method for processing features in parallel Vision Mamba, named PVM Layer, which achieves excellent performance with the lowest computational load while keeping the overall number of processing channels constant. We conducted comparisons and ablation experiments with several state-of-the-art lightweight models on three skin lesion public datasets and demonstrated that the UltraLight VM-UNet exhibits the same strong performance competitiveness with parameters of only 0.049M and GFLOPs of 0.060. In addition, this study deeply explores the key elements of parameter influence in Mamba, which will lay a theoretical foundation for Mamba to possibly become a new mainstream module for lightweighting in the future. The code is available from https://github.com/wurenkai/UltraLight-VM-UNet .
Authors: Bowen Lei, Dongkuan Xu, Ruqi Zhang, Bani Mallick
Abstract: Sparse training has emerged as a promising method for resource-efficient deep neural networks (DNNs) in real-world applications. However, the reliability of sparse models remains a crucial concern, particularly in detecting unknown out-of-distribution (OOD) data. This study addresses the knowledge gap by investigating the reliability of sparse training from an OOD perspective and reveals that sparse training exacerbates OOD unreliability. The lack of unknown information and the sparse constraints hinder the effective exploration of weight space and accurate differentiation between known and unknown knowledge. To tackle these challenges, we propose a new unknown-aware sparse training method, which incorporates a loss modification, auto-tuning strategy, and a voting scheme to guide weight space exploration and mitigate confusion between known and unknown information without incurring significant additional costs or requiring access to additional OOD data. Theoretical insights demonstrate how our method reduces model confidence when faced with OOD samples. Empirical experiments across multiple datasets, model architectures, and sparsity levels validate the effectiveness of our method, with improvements of up to \textbf{8.4\%} in AUROC while maintaining comparable or higher accuracy and calibration. This research enhances the understanding and readiness of sparse DNNs for deployment in resource-limited applications. Our code is available on: \url{https://github.com/StevenBoys/MOON}.
Authors: Luoyu Wang, Yitian Tao, Qing Yang, Yan Liang, Siwei Liu, Hongcheng Shi, Dinggang Shen, Han Zhang
Abstract: Simultaneous functional PET/MR (sf-PET/MR) presents a cutting-edge multimodal neuroimaging technique. It provides an unprecedented opportunity for concurrently monitoring and integrating multifaceted brain networks built by spatiotemporally covaried metabolic activity, neural activity, and cerebral blood flow (perfusion). Albeit high scientific/clinical values, short in hardware accessibility of PET/MR hinders its applications, let alone modern AI-based PET/MR fusion models. Our objective is to develop a clinically feasible AI-based disease diagnosis model trained on comprehensive sf-PET/MR data with the power of, during inferencing, allowing single modality input (e.g., PET only) as well as enforcing multimodal-based accuracy. To this end, we propose MX-ARM, a multimodal MiXture-of-experts Alignment and Reconstruction Model. It is modality detachable and exchangeable, allocating different multi-layer perceptrons dynamically ("mixture of experts") through learnable weights to learn respective representations from different modalities. Such design will not sacrifice model performance in uni-modal situation. To fully exploit the inherent complex and nonlinear relation among modalities while producing fine-grained representations for uni-modal inference, we subsequently add a modal alignment module to line up a dominant modality (e.g., PET) with representations of auxiliary modalities (MR). We further adopt multimodal reconstruction to promote the quality of learned features. Experiments on precious multimodal sf-PET/MR data for Mild Cognitive Impairment diagnosis showcase the efficacy of our model toward clinically feasible precision medicine.
Authors: Benjamin Townsend, Madison May, Christopher Wells
Abstract: We introduce RealKIE, a benchmark of five challenging datasets aimed at advancing key information extraction methods, with an emphasis on enterprise applications. The datasets include a diverse range of documents including SEC S1 Filings, US Non-disclosure Agreements, UK Charity Reports, FCC Invoices, and Resource Contracts. Each presents unique challenges: poor text serialization, sparse annotations in long documents, and complex tabular layouts. These datasets provide a realistic testing ground for key information extraction tasks like investment analysis and legal data processing. In addition to presenting these datasets, we offer an in-depth description of the annotation process, document processing techniques, and baseline modeling approaches. This contribution facilitates the development of NLP models capable of handling practical challenges and supports further research into information extraction technologies applicable to industry-specific problems. The annotated data and OCR outputs are available to download at https://indicodatasolutions.github.io/RealKIE/ code to reproduce the baselines will be available shortly.
Authors: Chuan Huang, Jia Wei, Rui Li
Abstract: Multi-modal brain images from MRI scans are widely used in clinical diagnosis to provide complementary information from different modalities. However, obtaining fully paired multi-modal images in practice is challenging due to various factors, such as time, cost, and artifacts, resulting in modality-missing brain images. To address this problem, unsupervised multi-modal brain image translation has been extensively studied. Existing methods suffer from the problem of brain tumor deformation during translation, as they fail to focus on the tumor areas when translating the whole images. In this paper, we propose an unsupervised tumor-aware distillation teacher-student network called UTAD-Net, which is capable of perceiving and translating tumor areas precisely. Specifically, our model consists of two parts: a teacher network and a student network. The teacher network learns an end-to-end mapping from source to target modality using unpaired images and corresponding tumor masks first. Then, the translation knowledge is distilled into the student network, enabling it to generate more realistic tumor areas and whole images without masks. Experiments show that our model achieves competitive performance on both quantitative and qualitative evaluations of image quality compared with state-of-the-art methods. Furthermore, we demonstrate the effectiveness of the generated images on downstream segmentation tasks. Our code is available at https://github.com/scut-HC/UTAD-Net.
Authors: Yang You, Wenhao He, Jin Liu, Hongkai Xiong, Weiming Wang, Cewu Lu
Abstract: Object pose estimation constitutes a critical area within the domain of 3D vision. While contemporary state-of-the-art methods that leverage real-world pose annotations have demonstrated commendable performance, the procurement of such real training data incurs substantial costs. This paper focuses on a specific setting wherein only 3D CAD models are utilized as a priori knowledge, devoid of any background or clutter information. We introduce a novel method, CPPF++, designed for sim-to-real pose estimation. This method builds upon the foundational point-pair voting scheme of CPPF, reformulating it through a probabilistic view. To address the challenge posed by vote collision, we propose a novel approach that involves modeling the voting uncertainty by estimating the probabilistic distribution of each point pair within the canonical space. Furthermore, we augment the contextual information provided by each voting unit through the introduction of N-point tuples. To enhance the robustness and accuracy of the model, we incorporate several innovative modules, including noisy pair filtering, online alignment optimization, and a tuple feature ensemble. Alongside these methodological advancements, we introduce a new category-level pose estimation dataset, named DiversePose 300. Empirical evidence demonstrates that our method significantly surpasses previous sim-to-real approaches and achieves comparable or superior performance on novel datasets. Our code is available on https://github.com/qq456cvb/CPPF2.
Authors: Young-Jun Lee, Byungsoo Ko, Han-Gyu Kim, Jonghwan Hyeon, Ho-Jin Choi
Abstract: As sharing images in an instant message is a crucial factor, there has been active research on learning an image-text multi-modal dialogue models. However, training a well-generalized multi-modal dialogue model remains challenging due to the low quality and limited diversity of images per dialogue in existing multi-modal dialogue datasets. In this paper, we propose an automated pipeline to construct a multi-modal dialogue dataset, ensuring both dialogue quality and image diversity without requiring minimum human effort. In our pipeline, to guarantee the coherence between images and dialogue, we prompt GPT-4 to infer potential image-sharing moments - specifically, the utterance, speaker, rationale, and image description. Furthermore, we leverage CLIP similarity to maintain consistency between aligned multiple images to the utterance. Through this pipeline, we introduce DialogCC, a high-quality and diverse multi-modal dialogue dataset that surpasses existing datasets in terms of quality and diversity in human evaluation. Our comprehensive experiments highlight that when multi-modal dialogue models are trained using our dataset, their generalization performance on unseen dialogue datasets is significantly enhanced. We make our source code and dataset publicly available.
Authors: Fei Kong, Xiyue Wang, Jinxi Xiang, Sen Yang, Xinran Wang, Meng Yue, Jun Zhang, Junhan Zhao, Xiao Han, Yuhan Dong, Biyue Zhu, Fang Wang, Yueping Liu
Abstract: Artificial intelligence (AI) holds significant promise in transforming medical imaging, enhancing diagnostics, and refining treatment strategies. However, the reliance on extensive multicenter datasets for training AI models poses challenges due to privacy concerns. Federated learning provides a solution by facilitating collaborative model training across multiple centers without sharing raw data. This study introduces a federated attention-consistent learning (FACL) framework to address challenges associated with large-scale pathological images and data heterogeneity. FACL enhances model generalization by maximizing attention consistency between local clients and the server model. To ensure privacy and validate robustness, we incorporated differential privacy by introducing noise during parameter transfer. We assessed the effectiveness of FACL in cancer diagnosis and Gleason grading tasks using 19,461 whole-slide images of prostate cancer from multiple centers. In the diagnosis task, FACL achieved an area under the curve (AUC) of 0.9718, outperforming seven centers with an average AUC of 0.9499 when categories are relatively balanced. For the Gleason grading task, FACL attained a Kappa score of 0.8463, surpassing the average Kappa score of 0.7379 from six centers. In conclusion, FACL offers a robust, accurate, and cost-effective AI training model for prostate cancer pathology while maintaining effective data safeguards.
Authors: Yizhak Ben-Shabat, Oren Shrout, Stephen Gould
Abstract: We propose a novel method for 3D point cloud action recognition. Understanding human actions in RGB videos has been widely studied in recent years, however, its 3D point cloud counterpart remains under-explored. This is mostly due to the inherent limitation of the point cloud data modality -- lack of structure, permutation invariance, and varying number of points -- which makes it difficult to learn a spatio-temporal representation. To address this limitation, we propose the 3DinAction pipeline that first estimates patches moving in time (t-patches) as a key building block, alongside a hierarchical architecture that learns an informative spatio-temporal representation. We show that our method achieves improved performance on existing datasets, including DFAUST and IKEA ASM. Code is publicly available at https://github.com/sitzikbs/3dincaction.
Authors: Wenbo Hu, Hongjian Zhan, Xinchen Ma, Cong Liu, Bing Yin, Yue Lu
Abstract: In the field of historical manuscript research, scholars frequently encounter novel symbols in ancient texts, investing considerable effort in their identification and documentation. Although existing object detection methods achieve impressive performance on known categories, they struggle to recognize novel symbols without retraining. To address this limitation, we propose a Visually Guided Text Spotting (VGTS) approach that accurately spots novel characters using just one annotated support sample. The core of VGTS is a spatial alignment module consisting of a Dual Spatial Attention (DSA) block and a Geometric Matching (GM) block. The DSA block aims to identify, focus on, and learn discriminative spatial regions in the support and query images, mimicking the human visual spotting process. It first refines the support image by analyzing inter-channel relationships to identify critical areas, and then refines the query image by focusing on informative key points. The GM block, on the other hand, establishes the spatial correspondence between the two images, enabling accurate localization of the target character in the query image. To tackle the example imbalance problem in low-resource spotting tasks, we develop a novel torus loss function that enhances the discriminative power of the embedding space for distance metric learning. To further validate our approach, we introduce a new dataset featuring ancient Dongba hieroglyphics (DBH) associated with the Naxi minority of China. Extensive experiments on the DBH dataset and other public datasets, including EGY, VML-HD, TKH, and NC, show that VGTS consistently surpasses state-of-the-art methods. The proposed framework exhibits great potential for application in historical manuscript text spotting, enabling scholars to efficiently identify and document novel symbols with minimal annotation effort.
Authors: Kumara Kahatapitiya, Anurag Arnab, Arsha Nagrani, Michael S. Ryoo
Abstract: Vision-Language models (VLMs) have excelled in the image-domain -- especially in zero-shot settings -- thanks to the availability of vast pretraining data (i.e., paired image-text samples). However for videos, such paired data is not as abundant. Therefore, video-VLMs are usually designed by adapting pretrained image-VLMs to the video-domain, instead of training from scratch. All such recipes rely on augmenting visual embeddings with temporal information (i.e., image $\rightarrow$ video), often keeping text embeddings unchanged or even being discarded. In this paper, we argue the contrary, that better video-VLMs can be designed by focusing more on augmenting text, rather than visual information. More specifically, we introduce Video-conditioned Text Representations (VicTR): a form of text embeddings optimized w.r.t. visual embeddings, creating a more-flexible contrastive latent space. Our model can further make use of freely-available semantic information, in the form of visually-grounded auxiliary text (e.g. object or scene information). We evaluate our model on few-shot, zero-shot (HMDB-51, UCF-101), short-form (Kinetics-400) and long-form (Charades) activity recognition benchmarks, showing strong performance among video-VLMs.
Authors: Aniket Gupta, Yiming Xie, Hanumant Singh, Huaizu Jiang
Abstract: Deep neural networks endow the downsampled superpoints with highly discriminative feature representations. Previous dominant point cloud registration approaches match these feature representations as the first step, e.g., using the Sinkhorn algorithm. A RANSAC-like method is then usually adopted as a post-processing refinement to filter the outliers. Other dominant method is to directly predict the superpoint matchings using learned MLP layers. Both of them have drawbacks: RANSAC-based methods are computationally intensive and prediction-based methods suffer from outputing non-existing points in the point cloud. In this paper, we propose a straightforward and effective baseline to find correspondences of superpoints in a global matching manner. We employ the normalized matching scores as weights for each correspondence, allowing us to reject the outliers and further weigh the rest inliers when fitting the transformation matrix without relying on the cumbersome RANSAC. Moreover, the entire model can be trained in an end-to-end fashion, leading to better accuracy. Our simple yet effective baseline shows comparable or even better results than state-of-the-art methods on three datasets including ModelNet, 3DMatch, and KITTI. We do not advocate our approach to be \emph{the} solution for point cloud registration but use the results to emphasize the role of matching strategy for point cloud registration. The code and models are available at https://github.com/neu-vi/Superpoints_Registration.
Authors: Lukas Knobel, Tengda Han, Yuki M. Asano
Abstract: While recent supervised methods for reference-based object counting continue to improve the performance on benchmark datasets, they have to rely on small datasets due to the cost associated with manually annotating dozens of objects in images. We propose UnCounTR, a model that can learn this task without requiring any manual annotations. To this end, we construct "Self-Collages", images with various pasted objects as training samples, that provide a rich learning signal covering arbitrary object types and counts. Our method builds on existing unsupervised representations and segmentation techniques to successfully demonstrate for the first time the ability of reference-based counting without manual supervision. Our experiments show that our method not only outperforms simple baselines and generic models such as FasterRCNN and DETR, but also matches the performance of supervised counting models in some domains.
Authors: Isabel Funke, Dominik Rivoir, Stefanie Krell, Stefanie Speidel
Abstract: To enable context-aware computer assistance in the operating room of the future, cognitive systems need to understand automatically which surgical phase is being performed by the medical team. The primary source of information for surgical phase recognition is typically video, which presents two challenges: extracting meaningful features from the video stream and effectively modeling temporal information in the sequence of visual features. For temporal modeling, attention mechanisms have gained popularity due to their ability to capture long-range dependencies. In this paper, we explore design choices for attention in existing temporal models for surgical phase recognition and propose a novel approach that uses attention more effectively and does not require hand-crafted constraints: TUNeS, an efficient and simple temporal model that incorporates self-attention at the core of a convolutional U-Net structure. In addition, we propose to train the feature extractor, a standard CNN, together with an LSTM on preferably long video segments, i.e., with long temporal context. In our experiments, almost all temporal models performed better on top of feature extractors that were trained with longer temporal context. On these contextualized features, TUNeS achieves state-of-the-art results on the Cholec80 and AutoLaparo datasets.
Authors: Jiayi Ni, Senqiao Yang, Ran Xu, Jiaming Liu, Xiaoqi Li, Wenyu Jiao, Zehui Chen, Yi Liu, Shanghang Zhang
Abstract: Since autonomous driving systems usually face dynamic and ever-changing environments, continual test-time adaptation (CTTA) has been proposed as a strategy for transferring deployed models to continually changing target domains. However, the pursuit of long-term adaptation often introduces catastrophic forgetting and error accumulation problems, which impede the practical implementation of CTTA in the real world. Recently, existing CTTA methods mainly focus on utilizing a majority of parameters to fit target domain knowledge through self-training. Unfortunately, these approaches often amplify the challenge of error accumulation due to noisy pseudo-labels, and pose practical limitations stemming from the heavy computational costs associated with entire model updates. In this paper, we propose a distribution-aware tuning (DAT) method to make the semantic segmentation CTTA efficient and practical in real-world applications. DAT adaptively selects and updates two small groups of trainable parameters based on data distribution during the continual adaptation process, including domain-specific parameters (DSP) and task-relevant parameters (TRP). Specifically, DSP exhibits sensitivity to outputs with substantial distribution shifts, effectively mitigating the problem of error accumulation. In contrast, TRP are allocated to positions that are responsive to outputs with minor distribution shifts, which are fine-tuned to avoid the catastrophic forgetting problem. In addition, since CTTA is a temporal task, we introduce the Parameter Accumulation Update (PAU) strategy to collect the updated DSP and TRP in target domain sequences. We conduct extensive experiments on two widely-used semantic segmentation CTTA benchmarks, achieving promising performance compared to previous state-of-the-art methods.
Authors: Jiaxiang Tang, Jiawei Ren, Hang Zhou, Ziwei Liu, Gang Zeng
Abstract: Recent advances in 3D content creation mostly leverage optimization-based 3D generation via score distillation sampling (SDS). Though promising results have been exhibited, these methods often suffer from slow per-sample optimization, limiting their practical usage. In this paper, we propose DreamGaussian, a novel 3D content generation framework that achieves both efficiency and quality simultaneously. Our key insight is to design a generative 3D Gaussian Splatting model with companioned mesh extraction and texture refinement in UV space. In contrast to the occupancy pruning used in Neural Radiance Fields, we demonstrate that the progressive densification of 3D Gaussians converges significantly faster for 3D generative tasks. To further enhance the texture quality and facilitate downstream applications, we introduce an efficient algorithm to convert 3D Gaussians into textured meshes and apply a fine-tuning stage to refine the details. Extensive experiments demonstrate the superior efficiency and competitive generation quality of our proposed approach. Notably, DreamGaussian produces high-quality textured meshes in just 2 minutes from a single-view image, achieving approximately 10 times acceleration compared to existing methods.
Authors: Zhan Xiong, Junling He, Pieter Valkema, Tri Q. Nguyen, Maarten Naesens, Jesper Kers, Fons J. Verbeek
Abstract: Renal biopsies are the gold standard for diagnosis of kidney diseases. Lesion scores made by renal pathologists are semi-quantitative and exhibit high inter-observer variability. Automating lesion classification within segmented anatomical structures can provide decision support in quantification analysis and reduce the inter-observer variability. Nevertheless, classifying lesions in regions-of-interest (ROIs) is clinically challenging due to (a) a large amount of densely packed anatomical objects (up to 1000), (b) class imbalance across different compartments (at least 3), (c) significant variation in object scales (i.e. sizes and shapes), and (d) the presence of multi-label lesions per anatomical structure. Existing models lack the capacity to address these complexities efficiently and generically. This paper presents \textbf{a generalized technical solution} for large-scale, multi-source datasets with diverse lesions. Our approach utilizes two sub-networks: dense instance segmentation and lesion classification. We introduce \textbf{DiffRegFormer}, an end-to-end dense instance segmentation model designed for multi-class, multi-scale objects within ROIs. Combining diffusion models, transformers, and RCNNs, DiffRegFormer efficiently recognizes over 500 objects across three anatomical classes (glomeruli, tubuli, arteries) within ROIs on a single NVIDIA GeForce RTX 3090 GPU. On a dataset of 303 ROIs (from 148 Jones' silver-stained renal WSIs), it outperforms state of art models, achieving AP of 52.1\% (detection) and 46.8\% (segmentation). Our lesion classification sub-network achieves 89.2\% precision and 64.6\% recall on 21889 object patches (from the 303 ROIs). Importantly, the model demonstrates direct domain transfer to PAS-stained WSIs without fine-tuning.
Authors: Bohan Zhai, Shijia Yang, Chenfeng Xu, Sheng Shen, Kurt Keutzer, Chunyuan Li, Manling Li
Abstract: Current Large Multimodal Models (LMMs) achieve remarkable progress, yet there remains significant uncertainty regarding their ability to accurately apprehend visual details, that is, in performing detailed captioning. To address this, we introduce $\textit{CCEval}$, a GPT-4 assisted evaluation method for detailed captioning. Interestingly, while LMMs demonstrate minimal object existence hallucination in existing VQA benchmarks, our proposed evaluation reveals continued susceptibility to such hallucinations. In this paper, we make the first attempt to investigate such hallucination from different aspects, including image resolution, the language decoder size, and instruction data amount, quality, granularity. Our findings underscore the unwarranted inference when the language description includes details at a finer object granularity than what the vision module can ground or verify, thus inducing hallucination. To control such hallucinations, we further attribute the reliability of captioning to contextual knowledge (involving only contextually grounded objects) and parametric knowledge (containing inferred objects by the model). Thus, we introduce $\textit{HallE-Control}$, a controllable LMM in terms of $\textbf{Hall}$ucination in object $\textbf{E}$xistence. HallE-Control can condition the captioning to shift between (i) exclusively depicting contextual knowledge for grounded objects and (ii) blending it with parametric knowledge to imagine inferred objects. Our method reduces hallucination by 44% compared to LLaVA$_{7B}$ and maintains the object coverage.
Authors: Lijun Yu, Jos\'e Lezama, Nitesh B. Gundavarapu, Luca Versari, Kihyuk Sohn, David Minnen, Yong Cheng, Vighnesh Birodkar, Agrim Gupta, Xiuye Gu, Alexander G. Hauptmann, Boqing Gong, Ming-Hsuan Yang, Irfan Essa, David A. Ross, Lu Jiang
Abstract: While Large Language Models (LLMs) are the dominant models for generative tasks in language, they do not perform as well as diffusion models on image and video generation. To effectively use LLMs for visual generation, one crucial component is the visual tokenizer that maps pixel-space inputs to discrete tokens appropriate for LLM learning. In this paper, we introduce MAGVIT-v2, a video tokenizer designed to generate concise and expressive tokens for both videos and images using a common token vocabulary. Equipped with this new tokenizer, we show that LLMs outperform diffusion models on standard image and video generation benchmarks including ImageNet and Kinetics. In addition, we demonstrate that our tokenizer surpasses the previously top-performing video tokenizer on two more tasks: (1) video compression comparable to the next-generation video codec (VCC) according to human evaluations, and (2) learning effective representations for action recognition tasks.
Authors: Yossi Gandelsman, Alexei A. Efros, Jacob Steinhardt
Abstract: We investigate the CLIP image encoder by analyzing how individual model components affect the final representation. We decompose the image representation as a sum across individual image patches, model layers, and attention heads, and use CLIP's text representation to interpret the summands. Interpreting the attention heads, we characterize each head's role by automatically finding text representations that span its output space, which reveals property-specific roles for many heads (e.g. location or shape). Next, interpreting the image patches, we uncover an emergent spatial localization within CLIP. Finally, we use this understanding to remove spurious features from CLIP and to create a strong zero-shot image segmenter. Our results indicate that a scalable understanding of transformer models is attainable and can be used to repair and improve models.
Authors: Takeru Miyato, Bernhard Jaeger, Max Welling, Andreas Geiger
Abstract: As transformers are equivariant to the permutation of input tokens, encoding the positional information of tokens is necessary for many tasks. However, since existing positional encoding schemes have been initially designed for NLP tasks, their suitability for vision tasks, which typically exhibit different structural properties in their data, is questionable. We argue that existing positional encoding schemes are suboptimal for 3D vision tasks, as they do not respect their underlying 3D geometric structure. Based on this hypothesis, we propose a geometry-aware attention mechanism that encodes the geometric structure of tokens as relative transformation determined by the geometric relationship between queries and key-value pairs. By evaluating on multiple novel view synthesis (NVS) datasets in the sparse wide-baseline multi-view setting, we show that our attention, called Geometric Transform Attention (GTA), improves learning efficiency and performance of state-of-the-art transformer-based NVS models without any additional learned parameters and only minor computational overhead.
Authors: Sara Ghazanfari, Alexandre Araujo, Prashanth Krishnamurthy, Farshad Khorrami, Siddharth Garg
Abstract: Recent years have seen growing interest in developing and applying perceptual similarity metrics. Research has shown the superiority of perceptual metrics over pixel-wise metrics in aligning with human perception and serving as a proxy for the human visual system. On the other hand, as perceptual metrics rely on neural networks, there is a growing concern regarding their resilience, given the established vulnerability of neural networks to adversarial attacks. It is indeed logical to infer that perceptual metrics may inherit both the strengths and shortcomings of neural networks. In this work, we demonstrate the vulnerability of state-of-the-art perceptual similarity metrics based on an ensemble of ViT-based feature extractors to adversarial attacks. We then propose a framework to train a robust perceptual similarity metric called LipSim (Lipschitz Similarity Metric) with provable guarantees. By leveraging 1-Lipschitz neural networks as the backbone, LipSim provides guarded areas around each data point and certificates for all perturbations within an $\ell_2$ ball. Finally, a comprehensive set of experiments shows the performance of LipSim in terms of natural and certified scores and on the image retrieval application. The code is available at https://github.com/SaraGhazanfari/LipSim.
Authors: Sungjune Park, Hyunjun Kim, Yong Man Ro
Abstract: Large language models (LLMs) have shown their capabilities in understanding contextual and semantic information regarding knowledge of instance appearances. In this paper, we introduce a novel approach to utilize the strengths of LLMs in understanding contextual appearance variations and to leverage this knowledge into a vision model (here, pedestrian detection). While pedestrian detection is considered one of the crucial tasks directly related to our safety (e.g., intelligent driving systems), it is challenging because of varying appearances and poses in diverse scenes. Therefore, we propose to formulate language-derived appearance elements and incorporate them with visual cues in pedestrian detection. To this end, we establish a description corpus that includes numerous narratives describing various appearances of pedestrians and other instances. By feeding them through an LLM, we extract appearance knowledge sets that contain the representations of appearance variations. Subsequently, we perform a task-prompting process to obtain appearance elements which are guided representative appearance knowledge relevant to a downstream pedestrian detection task. The obtained knowledge elements are adaptable to various detection frameworks, so that we can provide plentiful appearance information by integrating the language-derived appearance elements with visual cues within a detector. Through comprehensive experiments with various pedestrian detectors, we verify the adaptability and effectiveness of our method showing noticeable performance gains and achieving state-of-the-art detection performance on two public pedestrian detection benchmarks (i.e., CrowdHuman and WiderPedestrian).
Authors: Jiezhang Cao, Yue Shi, Kai Zhang, Yulun Zhang, Radu Timofte, Luc Van Gool
Abstract: Diffusion model-based image restoration (IR) aims to use diffusion models to recover high-quality (HQ) images from degraded images, achieving promising performance. Due to the inherent property of diffusion models, most existing methods need long serial sampling chains to restore HQ images step-by-step, resulting in expensive sampling time and high computation costs. Moreover, such long sampling chains hinder understanding the relationship between inputs and restoration results since it is hard to compute the gradients in the whole chains. In this work, we aim to rethink the diffusion model-based IR models through a different perspective, i.e., a deep equilibrium (DEQ) fixed point system, called DeqIR. Specifically, we derive an analytical solution by modeling the entire sampling chain in these IR models as a joint multivariate fixed point system. Based on the analytical solution, we can conduct parallel sampling and restore HQ images without training. Furthermore, we compute fast gradients via DEQ inversion and found that initialization optimization can boost image quality and control the generation direction. Extensive experiments on benchmarks demonstrate the effectiveness of our method on typical IR tasks and real-world settings.
Authors: Christopher Liao, Theodoros Tsiligkaridis, Brian Kulis
Abstract: Over the past year, a large body of multimodal research has emerged around zero-shot evaluation using GPT descriptors. These studies boost the zero-shot accuracy of pretrained VL models with an ensemble of label-specific text generated by GPT. A recent study, WaffleCLIP, demonstrated that similar zero-shot accuracy can be achieved with an ensemble of random descriptors. However, both zero-shot methods are un-trainable and consequently sub-optimal when some few-shot out-of-distribution (OOD) training data is available. Inspired by these prior works, we present two more flexible methods called descriptor and word soups, which do not require an LLM at test time and can leverage training data to increase OOD target accuracy. Descriptor soup greedily selects a small set of textual descriptors using generic few-shot training data, then calculates robust class embeddings using the selected descriptors. Word soup greedily assembles a chain of words in a similar manner. Compared to existing few-shot soft prompt tuning methods, word soup requires fewer parameters by construction and less GPU memory, since it does not require backpropagation. Both soups outperform current published few-shot methods, even when combined with SoTA zero-shot methods, on cross-dataset and domain generalization benchmarks. Compared with SoTA prompt and descriptor ensembling methods, such as ProDA and WaffleCLIP, word soup achieves higher OOD accuracy with fewer ensemble members. Please checkout our code: github.com/Chris210634/word_soups
Authors: Lingchen Meng, Shiyi Lan, Hengduo Li, Jose M. Alvarez, Zuxuan Wu, Yu-Gang Jiang
Abstract: In-context segmentation aims at segmenting novel images using a few labeled example images, termed as "in-context examples", exploring content similarities between examples and the target. The resulting models can be generalized seamlessly to novel segmentation tasks, significantly reducing the labeling and training costs compared with conventional pipelines. However, in-context segmentation is more challenging than classic ones requiring the model to learn segmentation rules conditioned on a few samples. Unlike previous work with ad-hoc or non-end-to-end designs, we propose SEGIC, an end-to-end segment-in-context framework built upon a single vision foundation model (VFM). In particular, SEGIC leverages the emergent correspondence within VFM to capture dense relationships between target images and in-context samples. As such, information from in-context samples is then extracted into three types of instructions, i.e. geometric, visual, and meta instructions, serving as explicit conditions for the final mask prediction. SEGIC is a straightforward yet effective approach that yields state-of-the-art performance on one-shot segmentation benchmarks. Notably, SEGIC can be easily generalized to diverse tasks, including video object segmentation and open-vocabulary segmentation. Code will be available at https://github.com/MengLcool/SEGIC.
Authors: Weijie Li, Yang Wei, Tianpeng Liu, Yuenan Hou, Yuxuan Li, Zhen Liu, Yongxiang Liu, Li Liu
Abstract: The growing Synthetic Aperture Radar (SAR) data has the potential to build a foundation model through Self-Supervised Learning (SSL) methods, which can achieve various SAR Automatic Target Recognition (ATR) tasks with pre-training in large-scale unlabeled data and fine-tuning in small labeled samples. SSL aims to construct supervision signals directly from the data, which minimizes the need for expensive expert annotation and maximizes the use of the expanding data pool for a foundational model. This study investigates an effective SSL method for SAR ATR, which can pave the way for a foundation model in SAR ATR. The primary obstacles faced in SSL for SAR ATR are the small targets in remote sensing and speckle noise in SAR images, corresponding to the SSL approach and signals. To overcome these challenges, we present a novel Joint-Embedding Predictive Architecture for SAR ATR (SAR-JEPA), which leverages local masked patches to predict the multi-scale SAR gradient representations of unseen context. The key aspect of SAR-JEPA is integrating SAR domain features to ensure high-quality self-supervised signals as target features. Besides, we employ local masks and multi-scale features to accommodate the various small targets in remote sensing. By fine-tuning and evaluating our framework on three target recognition datasets (vehicle, ship, and aircraft) with four other datasets as pre-training, we demonstrate its outperformance over other SSL methods and its effectiveness with increasing SAR data. This study showcases the potential of SSL for SAR target recognition across diverse targets, scenes, and sensors.
Authors: Zongwei Wu, Jilai Zheng, Xiangxuan Ren, Florin-Alexandru Vasluianu, Chao Ma, Danda Pani Paudel, Luc Van Gool, Radu Timofte
Abstract: In the realm of video object tracking, auxiliary modalities such as depth, thermal, or event data have emerged as valuable assets to complement the RGB trackers. In practice, most existing RGB trackers learn a single set of parameters to use them across datasets and applications. However, a similar single-model unification for multi-modality tracking presents several challenges. These challenges stem from the inherent heterogeneity of inputs -- each with modality-specific representations, the scarcity of multi-modal datasets, and the absence of all the modalities at all times. In this work, we introduce Un-Track, a Unified Tracker of a single set of parameters for any modality. To handle any modality, our method learns their common latent space through low-rank factorization and reconstruction techniques. More importantly, we use only the RGB-X pairs to learn the common latent space. This unique shared representation seamlessly binds all modalities together, enabling effective unification and accommodating any missing modality, all within a single transformer-based architecture. Our Un-Track achieves +8.1 absolute F-score gain, on the DepthTrack dataset, by introducing only +2.14 (over 21.50) GFLOPs with +6.6M (over 93M) parameters, through a simple yet efficient prompting strategy. Extensive comparisons on five benchmark datasets with different modalities show that Un-Track surpasses both SOTA unified trackers and modality-specific counterparts, validating our effectiveness and practicality. The source code is publicly available at https://github.com/Zongwei97/UnTrack.
Authors: Yijun Yang, Tianyi Zhou, Kanxue Li, Dapeng Tao, Lusong Li, Li Shen, Xiaodong He, Jing Jiang, Yuhui Shi
Abstract: While large language models (LLMs) excel in a simulated world of texts, they struggle to interact with the more realistic world without perceptions of other modalities such as visual or audio signals. Although vision-language models (VLMs) integrate LLM modules (1) aligned with static image features, and (2) may possess prior knowledge of world dynamics (as demonstrated in the text world), they have not been trained in an embodied visual world and thus cannot align with its dynamics. On the other hand, training an embodied agent in a noisy visual world without expert guidance is often challenging and inefficient. In this paper, we train a VLM agent living in a visual world using an LLM agent excelling in a parallel text world. Specifically, we distill LLM's reflection outcomes (improved actions by analyzing mistakes) in a text world's tasks to finetune the VLM on the same tasks of the visual world, resulting in an Embodied Multi-Modal Agent (EMMA) quickly adapting to the visual world dynamics. Such cross-modality imitation learning between the two parallel worlds is achieved by a novel DAgger-DPO algorithm, enabling EMMA to generalize to a broad scope of new tasks without any further guidance from the LLM expert. Extensive evaluations on the ALFWorld benchmark's diverse tasks highlight EMMA's superior performance to SOTA VLM-based agents, e.g., 20%-70% improvement in the success rate.
Authors: Chancharik Mitra, Brandon Huang, Trevor Darrell, Roei Herzig
Abstract: The combination of strong visual backbones and Large Language Model (LLM) reasoning has led to Large Multimodal Models (LMMs) becoming the current standard for a wide range of vision and language (VL) tasks. However, recent research has shown that even the most advanced LMMs still struggle to capture aspects of compositional visual reasoning, such as attributes and relationships between objects. One solution is to utilize scene graphs (SGs)--a formalization of objects and their relations and attributes that has been extensively used as a bridge between the visual and textual domains. Yet, scene graph data requires scene graph annotations, which are expensive to collect and thus not easily scalable. Moreover, finetuning an LMM based on SG data can lead to catastrophic forgetting of the pretraining objective. To overcome this, inspired by chain-of-thought methods, we propose Compositional Chain-of-Thought (CCoT), a novel zero-shot Chain-of-Thought prompting method that utilizes SG representations in order to extract compositional knowledge from an LMM. Specifically, we first generate an SG using the LMM, and then use that SG in the prompt to produce a response. Through extensive experiments, we find that the proposed CCoT approach not only improves LMM performance on several vision and language VL compositional benchmarks but also improves the performance of several popular LMMs on general multimodal benchmarks, without the need for fine-tuning or annotated ground-truth SGs. Code: https://github.com/chancharikmitra/CCoT
Authors: Jiayun Luo, Siddhesh Khandelwal, Leonid Sigal, Boyang Li
Abstract: From image-text pairs, large-scale vision-language models (VLMs) learn to implicitly associate image regions with words, which prove effective for tasks like visual question answering. However, leveraging the learned association for open-vocabulary semantic segmentation remains a challenge. In this paper, we propose a simple, yet extremely effective, training-free technique, Plug-and-Play Open-Vocabulary Semantic Segmentation (PnP-OVSS) for this task. PnP-OVSS leverages a VLM with direct text-to-image cross-attention and an image-text matching loss. To balance between over-segmentation and under-segmentation, we introduce Salience Dropout; by iteratively dropping patches that the model is most attentive to, we are able to better resolve the entire extent of the segmentation mask. \shortname{} does not require any neural network training and performs hyperparameter tuning without the need for any segmentation annotations, even for a validation set. PnP-OVSS demonstrates substantial improvements over comparable baselines (+29.4% mIoU on Pascal VOC, +13.2% mIoU on Pascal Context, +14.0% mIoU on MS COCO, and +11.4% mIoU on ADE-20K.) and even outperforms most baselines that conduct additional network training on top of pretrained VLMs. Our codebase is at https://github.com/letitiabanana/PnP-OVSS.
Authors: Dai Shi
Abstract: Due to the depth degradation effect in residual connections, many efficient Vision Transformers models that rely on stacking layers for information exchange often fail to form sufficient information mixing, leading to unnatural visual perception. To address this issue, in this paper, we propose Aggregated Attention, a biomimetic design-based token mixer that simulates biological foveal vision and continuous eye movement while enabling each token on the feature map to have a global perception. Furthermore, we incorporate learnable tokens that interact with conventional queries and keys, which further diversifies the generation of affinity matrices beyond merely relying on the similarity between queries and keys. Our approach does not rely on stacking for information exchange, thus effectively avoiding depth degradation and achieving natural visual perception. Additionally, we propose Convolutional GLU, a channel mixer that bridges the gap between GLU and SE mechanism, which empowers each token to have channel attention based on its nearest neighbor image features, enhancing local modeling capability and model robustness. We combine aggregated attention and convolutional GLU to create a new visual backbone called TransNeXt. Extensive experiments demonstrate that our TransNeXt achieves state-of-the-art performance across multiple model sizes. At a resolution of $224^2$, TransNeXt-Tiny attains an ImageNet accuracy of 84.0%, surpassing ConvNeXt-B with 69% fewer parameters. Our TransNeXt-Base achieves an ImageNet accuracy of 86.2% and an ImageNet-A accuracy of 61.6% at a resolution of $384^2$, a COCO object detection mAP of 57.1, and an ADE20K semantic segmentation mIoU of 54.7.
Authors: Zhiwen Fan, Kevin Wang, Kairun Wen, Zehao Zhu, Dejia Xu, Zhangyang Wang
Abstract: Recent advancements in real-time neural rendering using point-based techniques have paved the way for the widespread adoption of 3D representations. However, foundational approaches like 3D Gaussian Splatting come with a substantial storage overhead caused by growing the SfM points to millions, often demanding gigabyte-level disk space for a single unbounded scene, posing significant scalability challenges and hindering the splatting efficiency. To address this challenge, we introduce LightGaussian, a novel method designed to transform 3D Gaussians into a more efficient and compact format. Drawing inspiration from the concept of Network Pruning, LightGaussian identifies Gaussians that are insignificant in contributing to the scene reconstruction and adopts a pruning and recovery process, effectively reducing redundancy in Gaussian counts while preserving visual effects. Additionally, LightGaussian employs distillation and pseudo-view augmentation to distill spherical harmonics to a lower degree, allowing knowledge transfer to more compact representations while maintaining reflectance. Furthermore, we propose a hybrid scheme, VecTree Quantization, to quantize all attributes, resulting in lower bitwidth representations with minimal accuracy losses. In summary, LightGaussian achieves an averaged compression rate over 15x while boosting the FPS from 139 to 215, enabling an efficient representation of complex scenes on Mip-NeRF 360, Tank and Temple datasets. Project website: https://lightgaussian.github.io/
Authors: Jo\~ao Carreira, Michael King, Viorica P\u{a}tr\u{a}ucean, Dilara Gokay, C\u{a}t\u{a}lin Ionescu, Yi Yang, Daniel Zoran, Joseph Heyward, Carl Doersch, Yusuf Aytar, Dima Damen, Andrew Zisserman
Abstract: We introduce a framework for online learning from a single continuous video stream -- the way people and animals learn, without mini-batches, data augmentation or shuffling. This poses great challenges given the high correlation between consecutive video frames and there is very little prior work on it. Our framework allows us to do a first deep dive into the topic and includes a collection of streams and tasks composed from two existing video datasets, plus methodology for performance evaluation that considers both adaptation and generalization. We employ pixel-to-pixel modelling as a practical and flexible way to switch between pre-training and single-stream evaluation as well as between arbitrary tasks, without ever requiring changes to models and always using the same pixel loss. Equipped with this framework we obtained large single-stream learning gains from pre-training with a novel family of future prediction tasks, found that momentum hurts, and that the pace of weight updates matters. The combination of these insights leads to matching the performance of IID learning with batch size 1, when using the same architecture and without costly replay buffers.
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: Baptiste Brument, Robin Bruneau, Yvain Qu\'eau, Jean M\'elou, Fran\c{c}ois Bernard Lauze, Jean-Denis, Jean-Denis Durou, Lilian Calvet
Abstract: This paper introduces a versatile paradigm for integrating multi-view reflectance (optional) and normal maps acquired through photometric stereo. Our approach employs a pixel-wise joint re-parameterization of reflectance and normal, considering them as a vector of radiances rendered under simulated, varying illumination. This re-parameterization enables the seamless integration of reflectance and normal maps as input data in neural volume rendering-based 3D reconstruction while preserving a single optimization objective. In contrast, recent multi-view photometric stereo (MVPS) methods depend on multiple, potentially conflicting objectives. Despite its apparent simplicity, our proposed approach outperforms state-of-the-art approaches in MVPS benchmarks across F-score, Chamfer distance, and mean angular error metrics. Notably, it significantly improves the detailed 3D reconstruction of areas with high curvature or low visibility.
Authors: Jun Xiang, Xuan Gao, Yudong Guo, Juyong Zhang
Abstract: We propose FlashAvatar, a novel and lightweight 3D animatable avatar representation that could reconstruct a digital avatar from a short monocular video sequence in minutes and render high-fidelity photo-realistic images at 300FPS on a consumer-grade GPU. To achieve this, we maintain a uniform 3D Gaussian field embedded in the surface of a parametric face model and learn extra spatial offset to model non-surface regions and subtle facial details. While full use of geometric priors can capture high-frequency facial details and preserve exaggerated expressions, proper initialization can help reduce the number of Gaussians, thus enabling super-fast rendering speed. Extensive experimental results demonstrate that FlashAvatar outperforms existing works regarding visual quality and personalized details and is almost an order of magnitude faster in rendering speed. Project page: https://ustc3dv.github.io/FlashAvatar/
Authors: Shijie Zhou, Haoran Chang, Sicheng Jiang, Zhiwen Fan, Zehao Zhu, Dejia Xu, Pradyumna Chari, Suya You, Zhangyang Wang, Achuta Kadambi
Abstract: 3D scene representations have gained immense popularity in recent years. Methods that use Neural Radiance fields are versatile for traditional tasks such as novel view synthesis. In recent times, some work has emerged that aims to extend the functionality of NeRF beyond view synthesis, for semantically aware tasks such as editing and segmentation using 3D feature field distillation from 2D foundation models. However, these methods have two major limitations: (a) they are limited by the rendering speed of NeRF pipelines, and (b) implicitly represented feature fields suffer from continuity artifacts reducing feature quality. Recently, 3D Gaussian Splatting has shown state-of-the-art performance on real-time radiance field rendering. In this work, we go one step further: in addition to radiance field rendering, we enable 3D Gaussian splatting on arbitrary-dimension semantic features via 2D foundation model distillation. This translation is not straightforward: naively incorporating feature fields in the 3DGS framework encounters significant challenges, notably the disparities in spatial resolution and channel consistency between RGB images and feature maps. We propose architectural and training changes to efficiently avert this problem. Our proposed method is general, and our experiments showcase novel view semantic segmentation, language-guided editing and segment anything through learning feature fields from state-of-the-art 2D foundation models such as SAM and CLIP-LSeg. Across experiments, our distillation method is able to provide comparable or better results, while being significantly faster to both train and render. Additionally, to the best of our knowledge, we are the first method to enable point and bounding-box prompting for radiance field manipulation, by leveraging the SAM model. Project website at: https://feature-3dgs.github.io/
Authors: Ying Wang, Yanlai Yang, Mengye Ren
Abstract: In this paper we introduce LifelongMemory, a new framework for accessing long-form egocentric videographic memory through natural language question answering and retrieval. LifelongMemory generates concise video activity descriptions of the camera wearer and leverages the zero-shot capabilities of pretrained large language models to perform reasoning over long-form video context. Furthermore, Lifelong Memory uses a confidence and explanation module to produce confident, high-quality, and interpretable answers. Our approach achieves state-of-the-art performance on the EgoSchema benchmark for question answering and is highly competitive on the natural language query (NLQ) challenge of Ego4D. Code is available at https://github.com/Agentic-Learning-AI-Lab/lifelong-memory.
URLs: https://github.com/Agentic-Learning-AI-Lab/lifelong-memory.
Authors: Mohammad Reza Taesiri, Tianjun Feng, Anh Nguyen, Cor-Paul Bezemer
Abstract: Large multimodal models (LMMs) have evolved from large language models (LLMs) to integrate multiple input modalities, such as visual inputs. This integration augments the capacity of LLMs for tasks requiring visual comprehension and reasoning. However, the extent and limitations of their enhanced abilities are not fully understood, especially when it comes to real-world tasks. To address this gap, we introduce GlitchBench, a novel benchmark derived from video game quality assurance tasks, to test and evaluate the reasoning capabilities of LMMs. Our benchmark is curated from a variety of unusual and glitched scenarios from video games and aims to challenge both the visual and linguistic reasoning powers of LMMs in detecting and interpreting out-of-the-ordinary events. We evaluate multiple state-of-the-art LMMs, and we show that GlitchBench presents a new challenge for these models. Code and data are available at: https://glitchbench.github.io/
Authors: Tianchen Deng, Guole Shen, Tong Qin, Jianyu Wang, Wentao Zhao, Jingchuan Wang, Danwei Wang, Weidong Chen
Abstract: Neural implicit scene representations have recently shown encouraging results in dense visual SLAM. However, existing methods produce low-quality scene reconstruction and low-accuracy localization performance when scaling up to large indoor scenes and long sequences. These limitations are mainly due to their single, global radiance field with finite capacity, which does not adapt to large scenarios. Their end-to-end pose networks are also not robust enough with the growth of cumulative errors in large scenes. To this end, we introduce PLGSLAM, a neural visual SLAM system capable of high-fidelity surface reconstruction and robust camera tracking in real-time. To handle large-scale indoor scenes, PLGSLAM proposes a progressive scene representation method which dynamically allocates new local scene representation trained with frames within a local sliding window. This allows us to scale up to larger indoor scenes and improves robustness (even under pose drifts). In local scene representation, PLGSLAM utilizes tri-planes for local high-frequency features with multi-layer perceptron (MLP) networks for the low-frequency feature, achieving smoothness and scene completion in unobserved areas. Moreover, we propose local-to-global bundle adjustment method with a global keyframe database to address the increased pose drifts on long sequences. Experimental results demonstrate that PLGSLAM achieves state-of-the-art scene reconstruction results and tracking performance across various datasets and scenarios (both in small and large-scale indoor environments).
Authors: Zhi Gao, Yuntao Du, Xintong Zhang, Xiaojian Ma, Wenjuan Han, Song-Chun Zhu, Qing Li
Abstract: Utilizing large language models (LLMs) to compose off-the-shelf visual tools represents a promising avenue of research for developing robust visual assistants capable of addressing diverse visual tasks. However, these methods often overlook the potential for continual learning, typically by freezing the utilized tools, thus limiting their adaptation to environments requiring new knowledge. To tackle this challenge, we propose CLOVA, a Closed-Loop Visual Assistant, which operates within a framework encompassing inference, reflection, and learning phases. During the inference phase, LLMs generate programs and execute corresponding tools to complete assigned tasks. In the reflection phase, a multimodal global-local reflection scheme analyzes human feedback to determine which tools require updating. Lastly, the learning phase employs three flexible approaches to automatically gather training data and introduces a novel prompt tuning scheme to update the tools, allowing CLOVA to efficiently acquire new knowledge. Experimental findings demonstrate that CLOVA surpasses existing tool-usage methods by 5% in visual question answering and multiple-image reasoning, by 10% in knowledge tagging, and by 20% in image editing. These results underscore the significance of the continual learning capability in general visual assistants.
Authors: Ye Yuan, Xueting Li, Yangyi Huang, Shalini De Mello, Koki Nagano, Jan Kautz, Umar Iqbal
Abstract: Gaussian splatting has emerged as a powerful 3D representation that harnesses the advantages of both explicit (mesh) and implicit (NeRF) 3D representations. In this paper, we seek to leverage Gaussian splatting to generate realistic animatable avatars from textual descriptions, addressing the limitations (e.g., flexibility and efficiency) imposed by mesh or NeRF-based representations. However, a naive application of Gaussian splatting cannot generate high-quality animatable avatars and suffers from learning instability; it also cannot capture fine avatar geometries and often leads to degenerate body parts. To tackle these problems, we first propose a primitive-based 3D Gaussian representation where Gaussians are defined inside pose-driven primitives to facilitate animation. Second, to stabilize and amortize the learning of millions of Gaussians, we propose to use neural implicit fields to predict the Gaussian attributes (e.g., colors). Finally, to capture fine avatar geometries and extract detailed meshes, we propose a novel SDF-based implicit mesh learning approach for 3D Gaussians that regularizes the underlying geometries and extracts highly detailed textured meshes. Our proposed method, GAvatar, enables the large-scale generation of diverse animatable avatars using only text prompts. GAvatar significantly surpasses existing methods in terms of both appearance and geometry quality, and achieves extremely fast rendering (100 fps) at 1K resolution.
Authors: Minh-Quan Viet Bui, Jongmin Park, Jihyong Oh, Munchurl Kim
Abstract: Neural Radiance Fields (NeRF), initially developed for static scenes, have inspired many video novel view synthesis techniques. However, the challenge for video view synthesis arises from motion blur, a consequence of object or camera movement during exposure, which hinders the precise synthesis of sharp spatio-temporal views. In response, we propose a novel dynamic deblurring NeRF framework for blurry monocular video, called DyBluRF, consisting of a Base Ray Initialization (BRI) stage and a Motion Decomposition-based Deblurring (MDD) stage. Our DyBluRF is the first that handles the novel view synthesis for blurry monocular video with a novel two-stage framework. In the BRI stage, we coarsely reconstruct dynamic 3D scenes and jointly initialize the base ray, which is further used to predict latent sharp rays, using the inaccurate camera pose information from the given blurry frames. In the MDD stage, we introduce a novel Incremental Latent Sharp-rays Prediction (ILSP) approach for the blurry monocular video frames by decomposing the latent sharp rays into global camera motion and local object motion components. We further propose two loss functions for effective geometry regularization and decomposition of static and dynamic scene components without any mask supervision. Experiments show that DyBluRF outperforms qualitatively and quantitatively the SOTA methods.
Authors: Elnatan Kadar, Guy Gilboa
Abstract: We propose a new way to explain and to visualize neural network classification through a decomposition-based explainable AI (DXAI). Instead of providing an explanation heatmap, our method yields a decomposition of the image into class-agnostic and class-distinct parts, with respect to the data and chosen classifier. Following a fundamental signal processing paradigm of analysis and synthesis, the original image is the sum of the decomposed parts. We thus obtain a radically different way of explaining classification. The class-agnostic part ideally is composed of all image features which do not posses class information, where the class-distinct part is its complementary. This new visualization can be more helpful and informative in certain scenarios, especially when the attributes are dense, global and additive in nature, for instance, when colors or textures are essential for class distinction. Code is available at https://github.com/dxai2024/dxai.
Authors: Xiao Pan, Zongxin Yang, Shuai Bai, Yi Yang
Abstract: In this paper, we focus on the One-shot Novel View Synthesis (O-NVS) task which targets synthesizing photo-realistic novel views given only one reference image per scene. Previous One-shot Generalizable Neural Radiance Fields (OG-NeRF) methods solve this task in an inference-time finetuning-free manner, yet suffer the blurry issue due to the encoder-only architecture that highly relies on the limited reference image. On the other hand, recent diffusion-based image-to-3d methods show vivid plausible results via distilling pre-trained 2D diffusion models into a 3D representation, yet require tedious per-scene optimization. Targeting these issues, we propose the GD$^2$-NeRF, a Generative Detail compensation framework via GAN and Diffusion that is both inference-time finetuning-free and with vivid plausible details. In detail, following a coarse-to-fine strategy, GD$^2$-NeRF is mainly composed of a One-stage Parallel Pipeline (OPP) and a 3D-consistent Detail Enhancer (Diff3DE). At the coarse stage, OPP first efficiently inserts the GAN model into the existing OG-NeRF pipeline for primarily relieving the blurry issue with in-distribution priors captured from the training dataset, achieving a good balance between sharpness (LPIPS, FID) and fidelity (PSNR, SSIM). Then, at the fine stage, Diff3DE further leverages the pre-trained image diffusion models to complement rich out-distribution details while maintaining decent 3D consistency. Extensive experiments on both the synthetic and real-world datasets show that GD$^2$-NeRF noticeably improves the details while without per-scene finetuning.
Authors: Xingyu Zhou, Leheng Zhang, Xiaorui Zhao, Keze Wang, Leida Li, Shuhang Gu
Abstract: Recently, Vision Transformer has achieved great success in recovering missing details in low-resolution sequences, i.e., the video super-resolution (VSR) task. Despite its superiority in VSR accuracy, the heavy computational burden as well as the large memory footprint hinder the deployment of Transformer-based VSR models on constrained devices. In this paper, we address the above issue by proposing a novel feature-level masked processing framework: VSR with Masked Intra and inter frame Attention (MIA-VSR). The core of MIA-VSR is leveraging feature-level temporal continuity between adjacent frames to reduce redundant computations and make more rational use of previously enhanced SR features. Concretely, we propose an intra-frame and inter-frame attention block which takes the respective roles of past features and input features into consideration and only exploits previously enhanced features to provide supplementary information. In addition, an adaptive block-wise mask prediction module is developed to skip unimportant computations according to feature similarity between adjacent frames. We conduct detailed ablation studies to validate our contributions and compare the proposed method with recent state-of-the-art VSR approaches. The experimental results demonstrate that MIA-VSR improves the memory and computation efficiency over state-of-the-art methods, without trading off PSNR accuracy. The code is available at https://github.com/LabShuHangGU/MIA-VSR.
Authors: Serdar Erisen
Abstract: Improving the efficiency of state-of-the-art methods in semantic segmentation requires overcoming the increasing computational cost as well as issues such as fusing semantic information from global and local contexts. Based on the recent success and problems that convolutional neural networks (CNNs) encounter in semantic segmentation, this research proposes an encoder-decoder architecture with a unique efficient residual network, Efficient-ResNet. Attention-boosting gates (AbGs) and attention-boosting modules (AbMs) are deployed by aiming to fuse the equivariant and feature-based semantic information with the equivalent sizes of the output of global context of the efficient residual network in the encoder. Respectively, the decoder network is developed with the additional attention-fusion networks (AfNs) inspired by AbM. AfNs are designed to improve the efficiency in the one-to-one conversion of the semantic information by deploying additional convolution layers in the decoder part. Our network is tested on the challenging CamVid and Cityscapes datasets, and the proposed methods reveal significant improvements on the residual networks. To the best of our knowledge, the developed network, SERNet-Former, achieves state-of-the-art results (84.62 % mean IoU) on CamVid dataset and challenging results (87.35 % mean IoU) on Cityscapes validation dataset.
Authors: Till Beemelmanns, Quan Zhang, Lutz Eckstein
Abstract: Multi-modal 3D object detection models for automated driving have demonstrated exceptional performance on computer vision benchmarks like nuScenes. However, their reliance on densely sampled LiDAR point clouds and meticulously calibrated sensor arrays poses challenges for real-world applications. Issues such as sensor misalignment, miscalibration, and disparate sampling frequencies lead to spatial and temporal misalignment in data from LiDAR and cameras. Additionally, the integrity of LiDAR and camera data is often compromised by adverse environmental conditions such as inclement weather, leading to occlusions and noise interference. To address this challenge, we introduce MultiCorrupt, a comprehensive benchmark designed to evaluate the robustness of multi-modal 3D object detectors against ten distinct types of corruptions. We evaluate five state-of-the-art multi-modal detectors on MultiCorrupt and analyze their performance in terms of their resistance ability. Our results show that existing methods exhibit varying degrees of robustness depending on the type of corruption and their fusion strategy. We provide insights into which multi-modal design choices make such models robust against certain perturbations. The dataset generation code and benchmark are open-sourced at https://github.com/ika-rwth-aachen/MultiCorrupt.
Authors: Kihong Kim, Haneol Lee, Jihye Park, Seyeon Kim, Kwanghee Lee, Seungryong Kim, Jaejun Yoo
Abstract: Generating high-quality videos that synthesize desired realistic content is a challenging task due to their intricate high-dimensionality and complexity of videos. Several recent diffusion-based methods have shown comparable performance by compressing videos to a lower-dimensional latent space, using traditional video autoencoder architecture. However, such method that employ standard frame-wise 2D and 3D convolution fail to fully exploit the spatio-temporal nature of videos. To address this issue, we propose a novel hybrid video diffusion model, called HVDM, which can capture spatio-temporal dependencies more effectively. The HVDM is trained by a hybrid video autoencoder which extracts a disentangled representation of the video including: (i) a global context information captured by a 2D projected latent (ii) a local volume information captured by 3D convolutions with wavelet decomposition (iii) a frequency information for improving the video reconstruction. Based on this disentangled representation, our hybrid autoencoder provide a more comprehensive video latent enriching the generated videos with fine structures and details. Experiments on video generation benchamarks (UCF101, SkyTimelapse, and TaiChi) demonstrate that the proposed approach achieves state-of-the-art video generation quality, showing a wide range of video applications (e.g., long video generation, image-to-video, and video dynamics control).
Authors: Yi Feng, Yu Ma, Qijun Chen, Ioannis Pitas, Rui Fan
Abstract: Feature-fusion networks with duplex encoders have proven to be an effective technique to solve the freespace detection problem. However, despite the compelling results achieved by previous research efforts, the exploration of adequate and discriminative heterogeneous feature fusion, as well as the development of fallibility-aware loss functions remains relatively scarce. This paper makes several significant contributions to address these limitations: (1) It presents a novel heterogeneous feature fusion block, comprising a holistic attention module, a heterogeneous feature contrast descriptor, and an affinity-weighted feature recalibrator, enabling a more in-depth exploitation of the inherent characteristics of the extracted features, (2) it incorporates both inter-scale and intra-scale skip connections into the decoder architecture while eliminating redundant ones, leading to both improved accuracy and computational efficiency, and (3) it introduces two fallibility-aware loss functions that separately focus on semantic-transition and depth-inconsistent regions, collectively contributing to greater supervision during model training. Our proposed heterogeneous feature fusion network (SNE-RoadSegV2), which incorporates all these innovative components, demonstrates superior performance in comparison to all other freespace detection algorithms across multiple public datasets. Notably, it ranks the 1st on the official KITTI Road benchmark.
Authors: Xi Liu, Ying Guo, Cheng Zhen, Tong Li, Yingying Ao, Pengfei Yan
Abstract: Listening head generation aims to synthesize a non-verbal responsive listener head by modeling the correlation between the speaker and the listener in dynamic conversion.The applications of listener agent generation in virtual interaction have promoted many works achieving the diverse and fine-grained motion generation. However, they can only manipulate motions through simple emotional labels, but cannot freely control the listener's motions. Since listener agents should have human-like attributes (e.g. identity, personality) which can be freely customized by users, this limits their realism. In this paper, we propose a user-friendly framework called CustomListener to realize the free-form text prior guided listener generation. To achieve speaker-listener coordination, we design a Static to Dynamic Portrait module (SDP), which interacts with speaker information to transform static text into dynamic portrait token with completion rhythm and amplitude information. To achieve coherence between segments, we design a Past Guided Generation Module (PGG) to maintain the consistency of customized listener attributes through the motion prior, and utilize a diffusion-based structure conditioned on the portrait token and the motion prior to realize the controllable generation. To train and evaluate our model, we have constructed two text-annotated listening head datasets based on ViCo and RealTalk, which provide text-video paired labels. Extensive experiments have verified the effectiveness of our model.
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: Pum Jun Kim, Seojun Kim, Jaejun Yoo
Abstract: Image generative models have made significant progress in generating realistic and diverse images, supported by comprehensive guidance from various evaluation metrics. However, current video generative models struggle to generate even short video clips, with limited tools that provide insights for improvements. Current video evaluation metrics are simple adaptations of image metrics by switching the embeddings with video embedding networks, which may underestimate the unique characteristics of video. Our analysis reveals that the widely used Frechet Video Distance (FVD) has a stronger emphasis on the spatial aspect than the temporal naturalness of video and is inherently constrained by the input size of the embedding networks used, limiting it to 16 frames. Additionally, it demonstrates considerable instability and diverges from human evaluations. To address the limitations, we propose STREAM, a new video evaluation metric uniquely designed to independently evaluate spatial and temporal aspects. This feature allows comprehensive analysis and evaluation of video generative models from various perspectives, unconstrained by video length. We provide analytical and experimental evidence demonstrating that STREAM provides an effective evaluation tool for both visual and temporal quality of videos, offering insights into area of improvement for video generative models. To the best of our knowledge, STREAM is the first evaluation metric that can separately assess the temporal and spatial aspects of videos. Our code is available at https://github.com/pro2nit/STREAM.
Authors: Zhou Jiang, Zhenxin Zhu, Pengfei Li, Huan-ang Gao, Tianyuan Yuan, Yongliang Shi, Hang Zhao, Hao Zhao
Abstract: Autonomous vehicles are gradually entering city roads today, with the help of high-definition maps (HDMaps). However, the reliance on HDMaps prevents autonomous vehicles from stepping into regions without this expensive digital infrastructure. This fact drives many researchers to study online HDMap generation algorithms, but the performance of these algorithms at far regions is still unsatisfying. We present P-MapNet, in which the letter P highlights the fact that we focus on incorporating map priors to improve model performance. Specifically, we exploit priors in both SDMap and HDMap. On one hand, we extract weakly aligned SDMap from OpenStreetMap, and encode it as an additional conditioning branch. Despite the misalignment challenge, our attention-based architecture adaptively attends to relevant SDMap skeletons and significantly improves performance. On the other hand, we exploit a masked autoencoder to capture the prior distribution of HDMap, which can serve as a refinement module to mitigate occlusions and artifacts. We benchmark on the nuScenes and Argoverse2 datasets. Through comprehensive experiments, we show that: (1) our SDMap prior can improve online map generation performance, using both rasterized (by up to $+18.73$ $\rm mIoU$) and vectorized (by up to $+8.50$ $\rm mAP$) output representations. (2) our HDMap prior can improve map perceptual metrics by up to $6.34\%$. (3) P-MapNet can be switched into different inference modes that covers different regions of the accuracy-efficiency trade-off landscape. (4) P-MapNet is a far-seeing solution that brings larger improvements on longer ranges. Codes and models are publicly available at https://jike5.github.io/P-MapNet.
Authors: Guanzhou Ke, Bo Wang, Xiaoli Wang, Shengfeng He
Abstract: Multi-view representation learning aims to derive robust representations that are both view-consistent and view-specific from diverse data sources. This paper presents an in-depth analysis of existing approaches in this domain, highlighting a commonly overlooked aspect: the redundancy between view-consistent and view-specific representations. To this end, we propose an innovative framework for multi-view representation learning, which incorporates a technique we term 'distilled disentangling'. Our method introduces the concept of masked cross-view prediction, enabling the extraction of compact, high-quality view-consistent representations from various sources without incurring extra computational overhead. Additionally, we develop a distilled disentangling module that efficiently filters out consistency-related information from multi-view representations, resulting in purer view-specific representations. This approach significantly reduces redundancy between view-consistent and view-specific representations, enhancing the overall efficiency of the learning process. Our empirical evaluations reveal that higher mask ratios substantially improve the quality of view-consistent representations. Moreover, we find that reducing the dimensionality of view-consistent representations relative to that of view-specific representations further refines the quality of the combined representations. Our code is accessible at: https://github.com/Guanzhou-Ke/MRDD.
Authors: Baolu Li, Jinlong Li, Xinyu Liu, Runsheng Xu, Zhengzhong Tu, Jiacheng Guo, Xiaopeng Li, Hongkai Yu
Abstract: Current LiDAR-based Vehicle-to-Everything (V2X) multi-agent perception systems have shown the significant success on 3D object detection. While these models perform well in the trained clean weather, they struggle in unseen adverse weather conditions with the real-world domain gap. In this paper, we propose a domain generalization approach, named V2X-DGW, for LiDAR-based 3D object detection on multi-agent perception system under adverse weather conditions. Not only in the clean weather does our research aim to ensure favorable multi-agent performance, but also in the unseen adverse weather conditions by learning only on the clean weather data. To advance research in this area, we have simulated the impact of three prevalent adverse weather conditions on two widely-used multi-agent datasets, resulting in the creation of two novel benchmark datasets: OPV2V-w and V2XSet-w. To this end, we first introduce the Adaptive Weather Augmentation (AWA) to mimic the unseen adverse weather conditions, and then propose two alignments for generalizable representation learning: Trust-region Weather-invariant Alignment (TWA) and Agent-aware Contrastive Alignment (ACA). Extensive experimental results demonstrate that our V2X-DGW achieved improvements in the unseen adverse weather conditions.
Authors: Xiaoben Li, Mancheng Meng, Ziyan Wu, Terrence Chen, Fan Yang, Dinggang Shen
Abstract: Multi-view 3D human pose estimation is naturally superior to single view one, benefiting from more comprehensive information provided by images of multiple views. The information includes camera poses, 2D/3D human poses, and 3D geometry. However, the accurate annotation of these information is hard to obtain, making it challenging to predict accurate 3D human pose from multi-view images. To deal with this issue, we propose a fully self-supervised framework, named cascaded multi-view aggregating network (CMANet), to construct a canonical parameter space to holistically integrate and exploit multi-view information. In our framework, the multi-view information is grouped into two categories: 1) intra-view information , 2) inter-view information. Accordingly, CMANet consists of two components: intra-view module (IRV) and inter-view module (IEV). IRV is used for extracting initial camera pose and 3D human pose of each view; IEV is to fuse complementary pose information and cross-view 3D geometry for a final 3D human pose. To facilitate the aggregation of the intra- and inter-view, we define a canonical parameter space, depicted by per-view camera pose and human pose and shape parameters ($\theta$ and $\beta$) of SMPL model, and propose a two-stage learning procedure. At first stage, IRV learns to estimate camera pose and view-dependent 3D human pose supervised by confident output of an off-the-shelf 2D keypoint detector. At second stage, IRV is frozen and IEV further refines the camera pose and optimizes the 3D human pose by implicitly encoding the cross-view complement and 3D geometry constraint, achieved by jointly fitting predicted multi-view 2D keypoints. The proposed framework, modules, and learning strategy are demonstrated to be effective by comprehensive experiments and CMANet is superior to state-of-the-art methods in extensive quantitative and qualitative analysis.
Authors: Elena Ryumina, Maxim Markitantov, Dmitry Ryumin, Heysem Kaya, Alexey Karpov
Abstract: This paper presents the results of the SUN team for the Compound Expressions Recognition Challenge of the 6th ABAW Competition. We propose a novel audio-visual method for compound expression recognition. Our method relies on emotion recognition models that fuse modalities at the emotion probability level, while decisions regarding the prediction of compound expressions are based on predefined rules. Notably, our method does not use any training data specific to the target task. Thus, the problem is a zero-shot classification task. The method is evaluated in multi-corpus training and cross-corpus validation setups. Using our proposed method is achieved an F1-score value equals to 22.01% on the C-EXPR-DB test subset. Our findings from the challenge demonstrate that the proposed method can potentially form a basis for developing intelligent tools for annotating audio-visual data in the context of human's basic and compound emotions.
Authors: Yihong Luo, Xiaolong Chen, Jing Tang
Abstract: We introduce YOSO, a novel generative model designed for rapid, scalable, and high-fidelity one-step image synthesis. This is achieved by integrating the diffusion process with GANs. Specifically, we smooth the distribution by the denoising generator itself, performing self-cooperative learning. We show that our method can serve as a one-step generation model training from scratch with competitive performance. Moreover, we show that our method can be extended to finetune pre-trained text-to-image diffusion for high-quality one-step text-to-image synthesis even with LoRA fine-tuning. In particular, we provide the first diffusion transformer that can generate images in one step trained on 512 resolution, with the capability of adapting to 1024 resolution without explicit training. Our code is provided at https://github.com/Luo-Yihong/YOSO.
Authors: Zaid Tasneem, Akshat Dave, Abhishek Singh, Kushagra Tiwary, Praneeth Vepakomma, Ashok Veeraraghavan, Ramesh Raskar
Abstract: Neural radiance fields (NeRFs) show potential for transforming images captured worldwide into immersive 3D visual experiences. However, most of this captured visual data remains siloed in our camera rolls as these images contain personal details. Even if made public, the problem of learning 3D representations of billions of scenes captured daily in a centralized manner is computationally intractable. Our approach, DecentNeRF, is the first attempt at decentralized, crowd-sourced NeRFs that require $\sim 10^4\times$ less server computing for a scene than a centralized approach. Instead of sending the raw data, our approach requires users to send a 3D representation, distributing the high computation cost of training centralized NeRFs between the users. It learns photorealistic scene representations by decomposing users' 3D views into personal and global NeRFs and a novel optimally weighted aggregation of only the latter. We validate the advantage of our approach to learn NeRFs with photorealism and minimal server computation cost on structured synthetic and real-world photo tourism datasets. We further analyze how secure aggregation of global NeRFs in DecentNeRF minimizes the undesired reconstruction of personal content by the server.
Authors: Zhonghua Zhai, Chen Ju, Jinsong Lan, Shuai Xiao
Abstract: In this work, we propose Cell Variational Information Bottleneck Network (cellVIB), a convolutional neural network using information bottleneck mechanism, which can be combined with the latest feedforward network architecture in an end-to-end training method. Our Cell Variational Information Bottleneck Network is constructed by stacking VIB cells, which generate feature maps with uncertainty. As layers going deeper, the regularization effect will gradually increase, instead of directly adding excessive regular constraints to the output layer of the model as in Deep VIB. Under each VIB cell, the feedforward process learns an independent mean term and an standard deviation term, and predicts the Gaussian distribution based on them. The feedback process is based on reparameterization trick for effective training. This work performs an extensive analysis on MNIST dataset to verify the effectiveness of each VIB cells, and provides an insightful analysis on how the VIB cells affect mutual information. Experiments conducted on CIFAR-10 also prove that our cellVIB is robust against noisy labels during training and against corrupted images during testing. Then, we validate our method on PACS dataset, whose results show that the VIB cells can significantly improve the generalization performance of the basic model. Finally, in a more complex representation learning task, face recognition, our network structure has also achieved very competitive results.
Authors: Mai A. Shaaban, Adnan Khan, Mohammad Yaqub
Abstract: Chest X-ray images are commonly used for predicting acute and chronic cardiopulmonary conditions, but efforts to integrate them with structured clinical data face challenges due to incomplete electronic health records (EHR). This paper introduces MedPromptX, the first model to integrate multimodal large language models (MLLMs), few-shot prompting (FP) and visual grounding (VG) to combine imagery with EHR data for chest X-ray diagnosis. A pre-trained MLLM is utilized to complement the missing EHR information, providing a comprehensive understanding of patients' medical history. Additionally, FP reduces the necessity for extensive training of MLLMs while effectively tackling the issue of hallucination. Nevertheless, the process of determining the optimal number of few-shot examples and selecting high-quality candidates can be burdensome, yet it profoundly influences model performance. Hence, we propose a new technique that dynamically refines few-shot data for real-time adjustment to new patient scenarios. Moreover, VG aids in focusing the model's attention on relevant regions of interest in X-ray images, enhancing the identification of abnormalities. We release MedPromptX-VQA, a new in-context visual question answering dataset encompassing interleaved image and EHR data derived from MIMIC-IV and MIMIC-CXR databases. Results demonstrate the SOTA performance of MedPromptX, achieving an 11% improvement in F1-score compared to the baselines. Code and data are available at https://github.com/BioMedIA-MBZUAI/MedPromptX
Authors: Han Wang, Yanjie Wang, Yongjie Ye, Yuxiang Nie, Can Huang
Abstract: Multi-modal Large Language Models (MLLMs) have demonstrated their ability to perceive objects in still images, but their application in video-related tasks, such as object tracking, remains understudied. This lack of exploration is primarily due to two key challenges. Firstly, extensive pretraining on large-scale video datasets is required to equip MLLMs with the capability to perceive objects across multiple frames and understand inter-frame relationships. Secondly, processing a large number of frames within the context window of Large Language Models (LLMs) can impose a significant computational burden. To address the first challenge, we introduce ElysiumTrack-1M, a large-scale video dataset supported for three tasks: Single Object Tracking (SOT), Referring Single Object Tracking (RSOT), and Video Referring Expression Generation (Video-REG). ElysiumTrack-1M contains 1.27 million annotated video frames with corresponding object boxes and descriptions. Leveraging this dataset, we conduct training of MLLMs and propose a token-compression model T-Selector to tackle the second challenge. Our proposed approach, Elysium: Exploring Object-level Perception in Videos via MLLM, is an end-to-end trainable MLLM that attempts to conduct object-level tasks in videos without requiring any additional plug-in or expert models. All codes and datasets are available at https://github.com/Hon-Wong/Elysium.
Authors: Zhixin Lai, Jing Wu, Suiyao Chen, Yucheng Zhou, Naira Hovakimyan
Abstract: In this study, we uncover the unexpected efficacy of residual-based large language models (LLMs) as part of encoders for biomedical imaging tasks, a domain traditionally devoid of language or textual data. The approach diverges from established methodologies by utilizing a frozen transformer block, extracted from pre-trained LLMs, as an innovative encoder layer for the direct processing of visual tokens. This strategy represents a significant departure from the standard multi-modal vision-language frameworks, which typically hinge on language-driven prompts and inputs. We found that these LLMs could boost performance across a spectrum of biomedical imaging applications, including both 2D and 3D visual classification tasks, serving as plug-and-play boosters. More interestingly, as a byproduct, we found that the proposed framework achieved superior performance, setting new state-of-the-art results on extensive, standardized datasets in MedMNIST-2D and 3D. Through this work, we aim to open new avenues for employing LLMs in biomedical imaging and enriching the understanding of their potential in this specialized domain.
Authors: Yusuke Takimoto, Hikari Takehara, Hiroyuki Sato, Zihao Zhu, Bo Zheng
Abstract: In the film and gaming industries, achieving a realistic hair appearance typically involves the use of strands originating from the scalp. However, reconstructing these strands from observed surface images of hair presents significant challenges. The difficulty in acquiring Ground Truth (GT) data has led state-of-the-art learning-based methods to rely on pre-training with manually prepared synthetic CG data. This process is not only labor-intensive and costly but also introduces complications due to the domain gap when compared to real-world data. In this study, we propose an optimization-based approach that eliminates the need for pre-training. Our method represents hair strands as line segments growing from the scalp and optimizes them using a novel differentiable rendering algorithm. To robustly optimize a substantial number of slender explicit geometries, we introduce 3D orientation estimation utilizing global optimization, strand initialization based on Laplace's equation, and reparameterization that leverages geometric connectivity and spatial proximity. Unlike existing optimization-based methods, our method is capable of reconstructing internal hair flow in an absolute direction. Our method exhibits robust and accurate inverse rendering, surpassing the quality of existing methods and significantly improving processing speed.
Authors: Qiuhong Shen, Xuanyu Yi, Zike Wu, Pan Zhou, Hanwang Zhang, Shuicheng Yan, Xinchao Wang
Abstract: We tackle the challenge of efficiently reconstructing a 3D asset from a single image with growing demands for automated 3D content creation pipelines. Previous methods primarily rely on Score Distillation Sampling (SDS) and Neural Radiance Fields (NeRF). Despite their significant success, these approaches encounter practical limitations due to lengthy optimization and considerable memory usage. In this report, we introduce Gamba, an end-to-end amortized 3D reconstruction model from single-view images, emphasizing two main insights: (1) 3D representation: leveraging a large number of 3D Gaussians for an efficient 3D Gaussian splatting process; (2) Backbone design: introducing a Mamba-based sequential network that facilitates context-dependent reasoning and linear scalability with the sequence (token) length, accommodating a substantial number of Gaussians. Gamba incorporates significant advancements in data preprocessing, regularization design, and training methodologies. We assessed Gamba against existing optimization-based and feed-forward 3D generation approaches using the real-world scanned OmniObject3D dataset. Here, Gamba demonstrates competitive generation capabilities, both qualitatively and quantitatively, while achieving remarkable speed, approximately 0.6 second on a single NVIDIA A100 GPU.
Authors: Yanting Wang, Hongye Fu, Wei Zou, Jinyuan Jia
Abstract: Different from a unimodal model whose input is from a single modality, the input (called multi-modal input) of a multi-modal model is from multiple modalities such as image, 3D points, audio, text, etc. Similar to unimodal models, many existing studies show that a multi-modal model is also vulnerable to adversarial perturbation, where an attacker could add small perturbation to all modalities of a multi-modal input such that the multi-modal model makes incorrect predictions for it. Existing certified defenses are mostly designed for unimodal models, which achieve sub-optimal certified robustness guarantees when extended to multi-modal models as shown in our experimental results. In our work, we propose MMCert, the first certified defense against adversarial attacks to a multi-modal model. We derive a lower bound on the performance of our MMCert under arbitrary adversarial attacks with bounded perturbations to both modalities (e.g., in the context of auto-driving, we bound the number of changed pixels in both RGB image and depth image). We evaluate our MMCert using two benchmark datasets: one for the multi-modal road segmentation task and the other for the multi-modal emotion recognition task. Moreover, we compare our MMCert with a state-of-the-art certified defense extended from unimodal models. Our experimental results show that our MMCert outperforms the baseline.
Authors: Ekkasit Pinyoanuntapong, Muhammad Usama Saleem, Pu Wang, Minwoo Lee, Srijan Das, Chen Chen
Abstract: Generating human motion from text has been dominated by denoising motion models either through diffusion or generative masking process. However, these models face great limitations in usability by requiring prior knowledge of the motion length. Conversely, autoregressive motion models address this limitation by adaptively predicting motion endpoints, at the cost of degraded generation quality and editing capabilities. To address these challenges, we propose Bidirectional Autoregressive Motion Model (BAMM), a novel text-to-motion generation framework. BAMM consists of two key components: (1) a motion tokenizer that transforms 3D human motion into discrete tokens in latent space, and (2) a masked self-attention transformer that autoregressively predicts randomly masked tokens via a hybrid attention masking strategy. By unifying generative masked modeling and autoregressive modeling, BAMM captures rich and bidirectional dependencies among motion tokens, while learning the probabilistic mapping from textual inputs to motion outputs with dynamically-adjusted motion sequence length. This feature enables BAMM to simultaneously achieving high-quality motion generation with enhanced usability and built-in motion editability. Extensive experiments on HumanML3D and KIT-ML datasets demonstrate that BAMM surpasses current state-of-the-art methods in both qualitative and quantitative measures. Our project page is available at https://github.com/exitudio/BAMM-page.
Authors: Zhengwei Fang, Rui Wang, Tao Huang, Liping Jing
Abstract: Strong adversarial examples are crucial for evaluating and enhancing the robustness of deep neural networks. However, the performance of popular attacks is usually sensitive, for instance, to minor image transformations, stemming from limited information -- typically only one input example, a handful of white-box source models, and undefined defense strategies. Hence, the crafted adversarial examples are prone to overfit the source model, which hampers their transferability to unknown architectures. In this paper, we propose an approach named Multiple Asymptotically Normal Distribution Attacks (MultiANDA) which explicitly characterize adversarial perturbations from a learned distribution. Specifically, we approximate the posterior distribution over the perturbations by taking advantage of the asymptotic normality property of stochastic gradient ascent (SGA), then employ the deep ensemble strategy as an effective proxy for Bayesian marginalization in this process, aiming to estimate a mixture of Gaussians that facilitates a more thorough exploration of the potential optimization space. The approximated posterior essentially describes the stationary distribution of SGA iterations, which captures the geometric information around the local optimum. Thus, MultiANDA allows drawing an unlimited number of adversarial perturbations for each input and reliably maintains the transferability. Our proposed method outperforms ten state-of-the-art black-box attacks on deep learning models with or without defenses through extensive experiments on seven normally trained and seven defense models.
Authors: Govind Mittal, Chinmay Hegde, Nasir Memon
Abstract: With the rise of AI-enabled Real-Time Deepfakes (RTDFs), the integrity of online video interactions has become a growing concern. RTDFs have now made it feasible to replace an imposter's face with their victim in live video interactions. Such advancement in deepfakes also coaxes detection to rise to the same standard. However, existing deepfake detection techniques are asynchronous and hence ill-suited for RTDFs. To bridge this gap, we propose a challenge-response approach that establishes authenticity in live settings. We focus on talking-head style video interaction and present a taxonomy of challenges that specifically target inherent limitations of RTDF generation pipelines. We evaluate representative examples from the taxonomy by collecting a unique dataset comprising eight challenges, which consistently and visibly degrades the quality of state-of-the-art deepfake generators. These results are corroborated both by humans and a new automated scoring function, leading to 88.6% and 80.1% AUC, respectively. The findings underscore the promising potential of challenge-response systems for explainable and scalable real-time deepfake detection in practical scenarios. We provide access to data and code at https://github.com/mittalgovind/GOTCHA-Deepfakes
Authors: Guillaume Sall\'e, Pierre-Henri Conze, Julien Bert, Nicolas Boussion, Dimitris Visvikis, Vincent Jaouen
Abstract: \textit{Objectives}: Data scarcity and domain shifts lead to biased training sets that do not accurately represent deployment conditions. A related practical problem is cross-modal image segmentation, where the objective is to segment unlabelled images using previously labelled datasets from other imaging modalities. \textit{Methods}: We propose a cross-modal segmentation method based on conventional image synthesis boosted by a new data augmentation technique called Generative Blending Augmentation (GBA). GBA leverages a SinGAN model to learn representative generative features from a single training image to diversify realistically tumor appearances. This way, we compensate for image synthesis errors, subsequently improving the generalization power of a downstream segmentation model. The proposed augmentation is further combined to an iterative self-training procedure leveraging pseudo labels at each pass. \textit{Results}: The proposed solution ranked first for vestibular schwannoma (VS) segmentation during the validation and test phases of the MICCAI CrossMoDA 2022 challenge, with best mean Dice similarity and average symmetric surface distance measures. \textit{Conclusion and significance}: Local contrast alteration of tumor appearances and iterative self-training with pseudo labels are likely to lead to performance improvements in a variety of segmentation contexts.
Authors: Qinghao Ye, Haiyang Xu, Guohai Xu, Jiabo Ye, Ming Yan, Yiyang Zhou, Junyang Wang, Anwen Hu, Pengcheng Shi, Yaya Shi, Chenliang Li, Yuanhong Xu, Hehong Chen, Junfeng Tian, Qi Qian, Ji Zhang, Fei Huang, Jingren Zhou
Abstract: Large language models (LLMs) have demonstrated impressive zero-shot abilities on a variety of open-ended tasks, while recent research has also explored the use of LLMs for multi-modal generation. In this study, we introduce mPLUG-Owl, a novel training paradigm that equips LLMs with multi-modal abilities through modularized learning of foundation LLM, a visual knowledge module, and a visual abstractor module. This approach can support multiple modalities and facilitate diverse unimodal and multimodal abilities through modality collaboration. The training paradigm of mPLUG-Owl involves a two-stage method for aligning image and text, which learns visual knowledge with the assistance of LLM while maintaining and even improving the generation abilities of LLM. In the first stage, the visual knowledge module and abstractor module are trained with a frozen LLM module to align the image and text. In the second stage, language-only and multi-modal supervised datasets are used to jointly fine-tune a low-rank adaption (LoRA) module on LLM and the abstractor module by freezing the visual knowledge module. We carefully build a visually-related instruction evaluation set OwlEval. Experimental results show that our model outperforms existing multi-modal models, demonstrating mPLUG-Owl's impressive instruction and visual understanding ability, multi-turn conversation ability, and knowledge reasoning ability. Besides, we observe some unexpected and exciting abilities such as multi-image correlation and scene text understanding, which makes it possible to leverage it for harder real scenarios, such as vision-only document comprehension. Our code, pre-trained model, instruction-tuned models, and evaluation set are available at https://github.com/X-PLUG/mPLUG-Owl. The online demo is available at https://www.modelscope.cn/studios/damo/mPLUG-Owl.
URLs: https://github.com/X-PLUG/mPLUG-Owl., https://www.modelscope.cn/studios/damo/mPLUG-Owl.
Authors: Abdul Rehman Khan, Asifullah Khan
Abstract: Since their emergence, Convolutional Neural Networks (CNNs) have made significant strides in medical image analysis. However, the local nature of the convolution operator may pose a limitation for capturing global and long-range interactions in CNNs. Recently, Transformers have gained popularity in the computer vision community and also in medical image segmentation due to their ability to process global features effectively. The scalability issues of the self-attention mechanism and lack of the CNN-like inductive bias may have limited their adoption. Therefore, hybrid Vision transformers (CNN-Transformer), exploiting the advantages of both Convolution and Self-attention Mechanisms, have gained importance. In this work, we present MaxViT-UNet, a new Encoder-Decoder based UNet type hybrid vision transformer (CNN-Transformer) for medical image segmentation. The proposed Hybrid Decoder is designed to harness the power of both the convolution and self-attention mechanisms at each decoding stage with a nominal memory and computational burden. The inclusion of multi-axis self-attention, within each decoder stage, significantly enhances the discriminating capacity between the object and background regions, thereby helping in improving the segmentation efficiency. In the Hybrid Decoder, a new block is also proposed. The fusion process commences by integrating the upsampled lower-level decoder features, obtained through transpose convolution, with the skip-connection features derived from the hybrid encoder. Subsequently, the fused features undergo refinement through the utilization of a multi-axis attention mechanism. The proposed decoder block is repeated multiple times to segment the nuclei regions progressively. Experimental results on MoNuSeg18 and MoNuSAC20 datasets demonstrate the effectiveness of the proposed technique.
Authors: Sabine Muzellec, Thomas Fel, Victor Boutin, L\'eo and\'eol, Rufin VanRullen, Thomas Serre
Abstract: Attribution methods correspond to a class of explainability methods (XAI) that aim to assess how individual inputs contribute to a model's decision-making process. We have identified a significant limitation in one type of attribution methods, known as "white-box" methods. Although highly efficient, these methods rely on a gradient signal that is often contaminated by high-frequency noise. To overcome this limitation, we introduce a new approach called "FORGrad". This simple method effectively filters out noise artifacts by using optimal cut-off frequencies tailored to the unique characteristics of each model architecture. Our findings show that FORGrad consistently enhances the performance of already existing white-box methods, enabling them to compete effectively with more accurate yet computationally demanding "black-box" methods. We anticipate that our research will foster broader adoption of simpler and more efficient white-box methods for explainability, offering a better balance between faithfulness and computational efficiency.
Authors: Yang Qiu, Aaron B. Wagner, Johannes Ball\'e, Lucas Theis
Abstract: We introduce a distortion measure for images, Wasserstein distortion, that simultaneously generalizes pixel-level fidelity on the one hand and realism or perceptual quality on the other. We show how Wasserstein distortion reduces to a pure fidelity constraint or a pure realism constraint under different parameter choices and discuss its metric properties. Pairs of images that are close under Wasserstein distortion illustrate its utility. In particular, we generate random textures that have high fidelity to a reference texture in one location of the image and smoothly transition to an independent realization of the texture as one moves away from this point. Wasserstein distortion attempts to generalize and unify prior work on texture generation, image realism and distortion, and models of the early human visual system, in the form of an optimizable metric in the mathematical sense.
Authors: Antoine Salmona, Julie Delon, Agn\`es Desolneux
Abstract: The Gromov-Wasserstein (GW) distance is frequently used in machine learning to compare distributions across distinct metric spaces. Despite its utility, it remains computationally intensive, especially for large-scale problems. Recently, a novel Wasserstein distance specifically tailored for Gaussian mixture models and known as MW (mixture Wasserstein) has been introduced by several authors. In scenarios where data exhibit clustering, this approach simplifies to a small-scale discrete optimal transport problem, which complexity depends solely on the number of Gaussian components in the GMMs. This paper aims to extend MW by introducing new Gromov-type distances. These distances are designed to be isometry-invariant in Euclidean spaces and are applicable for comparing GMMs across different dimensional spaces. Our first contribution is the Mixture Gromov Wasserstein distance (MGW), which can be viewed as a Gromovized version of MW. This new distance has a straightforward discrete formulation, making it highly efficient for estimating distances between GMMs in practical applications. To facilitate the derivation of a transport plan between GMMs, we present a second distance, the Embedded Wasserstein distance (EW). This distance turns out to be closely related to several recent alternatives to Gromov-Wasserstein. We show that EW can be adapted to derive a distance as well as optimal transportation plans between GMMs. We demonstrate the efficiency of these newly proposed distances on medium to large-scale problems, including shape matching and hyperspectral image color transfer.
Authors: Luca Lusnig, Asel Sagingalieva, Mikhail Surmach, Tatjana Protasevich, Ovidiu Michiu, Joseph McLoughlin, Christopher Mansell, Graziano de' Petris, Deborah Bonazza, Fabrizio Zanconati, Alexey Melnikov, Fabio Cavalli
Abstract: In the realm of liver transplantation, accurately determining hepatic steatosis levels is crucial. Recognizing the essential need for improved diagnostic precision, particularly for optimizing diagnosis time by swiftly handling easy-to-solve cases and allowing the expert time to focus on more complex cases, this study aims to develop cutting-edge algorithms that enhance the classification of liver biopsy images. Additionally, the challenge of maintaining data privacy arises when creating automated algorithmic solutions, as sharing patient data between hospitals is restricted, further complicating the development and validation process. This research tackles diagnostic accuracy by leveraging novel techniques from the rapidly evolving field of quantum machine learning, known for their superior generalization abilities. Concurrently, it addresses privacy concerns through the implementation of privacy-conscious collaborative machine learning with federated learning. We introduce a hybrid quantum neural network model that leverages real-world clinical data to assess non-alcoholic liver steatosis accurately. This model achieves an image classification accuracy of 97%, surpassing traditional methods by 1.8%. Moreover, by employing a federated learning approach that allows data from different clients to be shared while ensuring privacy, we maintain an accuracy rate exceeding 90%. This initiative marks a significant step towards a scalable, collaborative, efficient, and dependable computational framework that aids clinical pathologists in their daily diagnostic tasks.
Authors: Amr Gomaa, Bilal Mahdy, Niko Kleer, Antonio Kr\"uger
Abstract: Robotic-assisted surgical systems have demonstrated significant potential in enhancing surgical precision and minimizing human errors. However, existing systems lack the ability to accommodate the unique preferences and requirements of individual surgeons. Additionally, they primarily focus on general surgeries (e.g., laparoscopy) and are not suitable for highly precise microsurgeries, such as ophthalmic procedures. Thus, we propose a simulation-based image-guided approach for surgeon-centered autonomous agents that can adapt to the individual surgeon's skill level and preferred surgical techniques during ophthalmic cataract surgery. Our approach utilizes a simulated environment to train reinforcement and imitation learning agents guided by image data to perform all tasks of the incision phase of cataract surgery. By integrating the surgeon's actions and preferences into the training process with the surgeon-in-the-loop, our approach enables the robot to implicitly learn and adapt to the individual surgeon's unique approach through demonstrations. This results in a more intuitive and personalized surgical experience for the surgeon. Simultaneously, it ensures consistent performance for the autonomous robotic apprentice. We define and evaluate the effectiveness of our approach using our proposed metrics; and highlight the trade-off between a generic agent and a surgeon-centered adapted agent. Moreover, our approach has the potential to extend to other ophthalmic surgical procedures, opening the door to a new generation of surgeon-in-the-loop autonomous surgical robots. We provide an open-source simulation framework for future development and reproducibility.
Authors: Yufan Deng, Ruida Wang, Yuhao Zhang, Yu-Wing Tai, Chi-Keung Tang
Abstract: Video generation models have shown their superior ability to generate photo-realistic video. However, how to accurately control (or edit) the video remains a formidable challenge. The main issues are: 1) how to perform direct and accurate user control in editing; 2) how to execute editings like changing shape, expression, and layout without unsightly distortion and artifacts to the edited content; and 3) how to maintain spatio-temporal consistency of video after editing. To address the above issues, we propose DragVideo, a general drag-style video editing framework. Inspired by DragGAN, DragVideo addresses issues 1) and 2) by proposing the drag-style video latent optimization method which gives desired control by updating noisy video latent according to drag instructions through video-level drag objective function. We amend issue 3) by integrating the video diffusion model with sample-specific LoRA and Mutual Self-Attention in DragVideo to ensure the edited result is spatio-temporally consistent. We also present a series of testing examples for drag-style video editing and conduct extensive experiments across a wide array of challenging editing tasks, such as motion, skeleton editing, etc, underscoring DragVideo can edit video in an intuitive, faithful to the user's intention manner, with nearly unnoticeable distortion and artifacts, while maintaining spatio-temporal consistency. While traditional prompt-based video editing fails to do the former two and directly applying image drag editing fails in the last, DragVideo's versatility and generality are emphasized. Github link: https://github.com/RickySkywalker/DragVideo-Official.
Authors: Yichao Liang, Kevin Ellis, Jo\~ao Henriques
Abstract: Developing generalizable manipulation skills is a core challenge in embodied AI. This includes generalization across diverse task configurations, encompassing variations in object shape, density, friction coefficient, and external disturbances such as forces applied to the robot. Rapid Motor Adaptation (RMA) offers a promising solution to this challenge. It posits that essential hidden variables influencing an agent's task performance, such as object mass and shape, can be effectively inferred from the agent's action and proprioceptive history. Drawing inspiration from RMA in locomotion and in-hand rotation, we use depth perception to develop agents tailored for rapid motor adaptation in a variety of manipulation tasks. We evaluated our agents on four challenging tasks from the Maniskill2 benchmark, namely pick-and-place operations with hundreds of objects from the YCB and EGAD datasets, peg insertion with precise position and orientation, and operating a variety of faucets and handles, with customized environment variations. Empirical results demonstrate that our agents surpass state-of-the-art methods like automatic domain randomization and vision-based policies, obtaining better generalization performance and sample efficiency.
Authors: No\"el Vouitsis, Zhaoyan Liu, Satya Krishna Gorti, Valentin Villecroze, Jesse C. Cresswell, Guangwei Yu, Gabriel Loaiza-Ganem, Maksims Volkovs
Abstract: The goal of multimodal alignment is to learn a single latent space that is shared between multimodal inputs. The most powerful models in this space have been trained using massive datasets of paired inputs and large-scale computational resources, making them prohibitively expensive to train in many practical scenarios. We surmise that existing unimodal encoders pre-trained on large amounts of unimodal data should provide an effective bootstrap to create multimodal models from unimodal ones at much lower costs. We therefore propose FuseMix, a multimodal augmentation scheme that operates on the latent spaces of arbitrary pre-trained unimodal encoders. Using FuseMix for multimodal alignment, we achieve competitive performance -- and in certain cases outperform state-of-the art methods -- in both image-text and audio-text retrieval, with orders of magnitude less compute and data: for example, we outperform CLIP on the Flickr30K text-to-image retrieval task with $\sim \! 600\times$ fewer GPU days and $\sim \! 80\times$ fewer image-text pairs. Additionally, we show how our method can be applied to convert pre-trained text-to-image generative models into audio-to-image ones. Code is available at: https://github.com/layer6ai-labs/fusemix.
Authors: Ishak Ayad, Nicolas Larue, Ma\"i K. Nguyen
Abstract: Inverse problems span across diverse fields. In medical contexts, computed tomography (CT) plays a crucial role in reconstructing a patient's internal structure, presenting challenges due to artifacts caused by inherently ill-posed inverse problems. Previous research advanced image quality via post-processing and deep unrolling algorithms but faces challenges, such as extended convergence times with ultra-sparse data. Despite enhancements, resulting images often show significant artifacts, limiting their effectiveness for real-world diagnostic applications. We aim to explore deep second-order unrolling algorithms for solving imaging inverse problems, emphasizing their faster convergence and lower time complexity compared to common first-order methods like gradient descent. In this paper, we introduce QN-Mixer, an algorithm based on the quasi-Newton approach. We use learned parameters through the BFGS algorithm and introduce Incept-Mixer, an efficient neural architecture that serves as a non-local regularization term, capturing long-range dependencies within images. To address the computational demands typically associated with quasi-Newton algorithms that require full Hessian matrix computations, we present a memory-efficient alternative. Our approach intelligently downsamples gradient information, significantly reducing computational requirements while maintaining performance. The approach is validated through experiments on the sparse-view CT problem, involving various datasets and scanning protocols, and is compared with post-processing and deep unrolling state-of-the-art approaches. Our method outperforms existing approaches and achieves state-of-the-art performance in terms of SSIM and PSNR, all while reducing the number of unrolling iterations required.
Authors: Nishant Jain, Arun S. Suggala, Pradeep Shenoy
Abstract: Learned reweighting (LRW) approaches to supervised learning use an optimization criterion to assign weights for training instances, in order to maximize performance on a representative validation dataset. We pose and formalize the problem of optimized selection of the validation set used in LRW training, to improve classifier generalization. In particular, we show that using hard-to-classify instances in the validation set has both a theoretical connection to, and strong empirical evidence of generalization. We provide an efficient algorithm for training this meta-optimized model, as well as a simple train-twice heuristic for careful comparative study. We demonstrate that LRW with easy validation data performs consistently worse than LRW with hard validation data, establishing the validity of our meta-optimization problem. Our proposed algorithm outperforms a wide range of baselines on a range of datasets and domain shift challenges (Imagenet-1K, CIFAR-100, Clothing-1M, CAMELYON, WILDS, etc.), with ~1% gains using VIT-B on Imagenet. We also show that using naturally hard examples for validation (Imagenet-R / Imagenet-A) in LRW training for Imagenet improves performance on both clean and naturally hard test instances by 1-2%. Secondary analyses show that using hard validation data in an LRW framework improves margins on test data, hinting at the mechanism underlying our empirical gains. We believe this work opens up new research directions for the meta-optimization of meta-learning in a supervised learning context.
Authors: Yushan Huang, Josh Millar, Yuxuan Long, Yuchen Zhao, Hamed Haddadi
Abstract: The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on fine-tuning either the full base model or its last few layers to adapt to new data, while often neglecting energy costs. However, various types of data drift exist, and fine-tuning the full base model or the last few layers may not result in optimal performance in certain scenarios. We propose Target Block Fine-Tuning (TBFT), a low-energy adaptive personalization framework designed for resource-constrained devices. We categorize data drift and personalization into three types: input-level, feature-level, and output-level. For each type, we fine-tune different blocks of the model to achieve optimal performance with reduced energy costs. Specifically, input-, feature-, and output-level correspond to fine-tuning the front, middle, and rear blocks of the model. We evaluate TBFT on a ResNet model, three datasets, three different training sizes, and a Raspberry Pi. Compared with the $Block Avg$, where each block is fine-tuned individually and their performance improvements are averaged, TBFT exhibits an improvement in model accuracy by an average of 15.30% whilst saving 41.57% energy consumption on average compared with full fine-tuning.
Authors: Zirui Qiu, Hassan Rivaz, Yiming Xiao
Abstract: As deep learning has become the state-of-the-art for computer-assisted diagnosis, interpretability of the automatic decisions is crucial for clinical deployment. While various methods were proposed in this domain, visual attention maps of clinicians during radiological screening offer a unique asset to provide important insights and can potentially enhance the quality of computer-assisted diagnosis. With this paper, we introduce a novel deep-learning framework for joint disease diagnosis and prediction of corresponding visual saliency maps for chest X-ray scans. Specifically, we designed a novel dual-encoder multi-task UNet, which leverages both a DenseNet201 backbone and a Residual and Squeeze-and-Excitation block-based encoder to extract diverse features for saliency map prediction, and a multi-scale feature-fusion classifier to perform disease classification. To tackle the issue of asynchronous training schedules of individual tasks in multi-task learning, we proposed a multi-stage cooperative learning strategy, with contrastive learning for feature encoder pretraining to boost performance. Experiments show that our proposed method outperformed existing techniques for chest X-ray diagnosis and the quality of visual saliency map prediction.