Authors: Anusha Verma, Ghazal Ghajari, K M Tawsik Jawad, Dr. Hugh P. Salehi, Dr. Fathi Amsaad
Abstract: Maintaining patient safety and the safety of healthcare workers (HCWs) in hospitals and clinics highly depends on following the proper protocol for donning and taking off personal protective equipment (PPE). HCWs can benefit from a feedback system during the putting on and removal process because the process is cognitively demanding and errors are common. Centers for Disease Control and Prevention (CDC) provided guidelines for correct PPE use which should be followed. A real time object detection along with a unique sequencing algorithms are used to identify and determine the donning and doffing process in real time. The purpose of this technical research is two-fold: The user gets real time alert to the step they missed in the sequence if they don't follow the proper procedure during donning or doffing. Secondly, the use of tiny machine learning (yolov4-tiny) in embedded system architecture makes it feasible and cost-effective to deploy in different healthcare settings.
Authors: Wei Wang, Qing Li
Abstract: Computer vision (CV) is one of the most crucial fields in artificial intelligence. In recent years, a variety of deep learning models based on convolutional neural networks (CNNs) and Transformers have been designed to tackle diverse problems in CV. These algorithms have found practical applications in areas such as robotics and facial recognition. Despite the increasing power of current CV models, several fundamental questions remain unresolved: Why do CNNs require deep layers? What ensures the generalization ability of CNNs? Why do residual-based networks outperform fully convolutional networks like VGG? What is the fundamental difference between residual-based CNNs and Transformer-based networks? Why can CNNs utilize LoRA and pruning techniques? The root cause of these questions lies in the lack of a robust theoretical foundation for deep learning models in CV. To address these critical issues and techniques, we employ the Universal Approximation Theorem (UAT) to provide a theoretical basis for convolution- and Transformer-based models in CV. By doing so, we aim to elucidate these questions from a theoretical perspective.
Authors: Thalles Silva, Helio Pedrini, Ad\'in Ram\'irez Rivera
Abstract: This paper introduces a novel approach to improving the training stability of self-supervised learning (SSL) methods by leveraging a non-parametric memory of seen concepts. The proposed method involves augmenting a neural network with a memory component to stochastically compare current image views with previously encountered concepts. Additionally, we introduce stochastic memory blocks to regularize training and enforce consistency between image views. We extensively benchmark our method on many vision tasks, such as linear probing, transfer learning, low-shot classification, and image retrieval on many datasets. The experimental results consolidate the effectiveness of the proposed approach in achieving stable SSL training without additional regularizers while learning highly transferable representations and requiring less computing time and resources.
Authors: Changdae Oh, Gyeongdeok Seo, Geunyoung Jung, Zhi-Qi Cheng, Hosik Choi, Jiyoung Jung, Kyungwoo Song
Abstract: With the surge of large-scale pre-trained models (PTMs), adapting these models to numerous downstream tasks becomes a crucial problem. Consequently, parameter-efficient transfer learning (PETL) of large models has grasped huge attention. While PETL methods show impressive performance, they commonly rely on two optimistic assumptions: 1) the entire parameters of a PTM are available, and 2) a sufficiently large memory capacity is equipped for caching all the intermediate activations to compute gradients. However, in most real-world applications, PTMs are served as black-box APIs or proprietary software without explicit parameter accessibility. Besides, it is hard to meet a large memory requirement for modern PTMs. This work proposes black-box visual prompting (BlackVIP), which efficiently adapts the PTMs without knowledge about model architectures and parameters. BlackVIP has two components; 1) Coordinator and 2) simultaneous perturbation stochastic approximation with gradient correction (SPSA-GC). The Coordinator designs input-dependent visual prompts, which allow the target PTM to adapt in the wild. SPSA-GC efficiently estimates the gradient of PTM to update the Coordinator. Besides, we propose a variant, BlackVIP-SE, which significantly reduces the runtime and computational cost of BlackVIP. Extensive experiments on 19 datasets demonstrate that BlackVIPs enable robust adaptation to diverse domains and tasks with minimal memory requirements. We further provide theoretical analysis on the generalization of visual prompting methods by presenting their connection to the certified robustness of randomized smoothing.
Authors: Youngseok Yoon, Dainong Hu, Iain Weissburg, Yao Qin, Haewon Jeong
Abstract: Diffusion models have achieved tremendous improvements in generative modeling for images, enabling high-quality generation that is indistinguishable by humans from real images. The qualities of images have reached a threshold at which we can reuse synthetic images for training machine learning models again. This attracts the area as it can relieve the high cost of data collection and fundamentally solve many problems in data-limited areas. In this paper, we focus on a practical scenario in which pretrained text-to-image diffusion models are iteratively finetuned using a set of synthetic images, which we call the Chain of Diffusion. Finetuned models generate images that are used for the next iteration of finetuning. We first demonstrate how these iterative processes result in severe degradation in image qualities. Thorough investigations reveal the most impactful factor for the degradation, and we propose finetuning and generation strategies that can effectively resolve the degradation. Our method, Reusable Diffusion Finetuning (ReDiFine), combines condition drop finetuning and CFG scheduling to maintain the qualities of generated images throughout iterations. ReDiFine works effectively for multiple datasets and models without further hyperparameter search, making synthetic images reusable to finetune future generative models.
Authors: Akanksha Dixit, Smruti R. Sarangi
Abstract: High-refresh rate displays have become very popular in recent years due to the need for superior visual quality in gaming, professional displays and specialized applications like medical imaging. However, high-refresh rate displays alone do not guarantee a superior visual experience; the GPU needs to render frames at a matching rate. Otherwise, we observe disconcerting visual artifacts such as screen tearing and stuttering. Temporal supersampling is an effective technique to increase frame rates by predicting new frames from other rendered frames. There are two methods in this space: interpolation and extrapolation. Interpolation-based methods provide good image quality at the cost of a higher latency because they also require the next rendered frame. On the other hand, extrapolation methods are much faster at the cost of quality. This paper introduces PatchEX, a novel frame extrapolation method that aims to provide the quality of interpolation at the speed of extrapolation. It smartly partitions the extrapolation task into sub-tasks and executes them in parallel to improve both quality and latency. It then uses a patch-based inpainting method and a custom shadow prediction approach to fuse the generated sub-frames. This approach significantly reduces the overall latency while maintaining the quality of the output. Our results demonstrate that PatchEX achieves a 65.29% and 48.46% improvement in PSNR over the latest extrapolation methods ExtraNet and ExtraSS, respectively, while being 6x and 2x faster, respectively.
Authors: Edoardo Cittadini, Alessandro De Siena, Giorgio Buttazzo
Abstract: The ever-increasing use of artificial intelligence in autonomous systems has significantly contributed to advance the research on multi-object tracking, adopted in several real-time applications (e.g., autonomous driving, surveillance drones, robotics) to localize and follow the trajectory of multiple objects moving in front of a camera. Current tracking algorithms can be divided into two main categories: some approaches introduce complex heuristics and re-identification models to improve the tracking accuracy and reduce the number of identification switches, without particular attention to the timing performance, whereas other approaches are aimed at reducing response times by removing the re-identification phase, thus penalizing the tracking accuracy. This work proposes a new approach to multi-class object tracking that allows achieving smaller and more predictable execution times, without penalizing the tracking performance. The idea is to reduce the problem of matching predictions with detections into smaller sub-problems by splitting the Hungarian matrix by class and invoking the second re-identification stage only when strictly necessary for a smaller number of elements. The proposed solution was evaluated in complex urban scenarios with several objects of different types (as cars, trucks, bikes, and pedestrians), showing the effectiveness of the multi-class approach with respect to state of the art trackers.
Authors: Hazem Hesham Yousef Shalby, Massimo Pavan, Manuel Roveri
Abstract: Tiny Machine Learning (TinyML) is a branch of Machine Learning (ML) that constitutes a bridge between the ML world and the embedded system ecosystem (i.e., Internet of Things devices, embedded devices, and edge computing units), enabling the execution of ML algorithms on devices constrained in terms of memory, computational capabilities, and power consumption. Video Streaming Analysis (VSA), one of the most interesting tasks of TinyML, consists in scanning a sequence of frames in a streaming manner, with the goal of identifying interesting patterns. Given the strict constraints of these tiny devices, all the current solutions rely on performing a frame-by-frame analysis, hence not exploiting the temporal component in the stream of data. In this paper, we present StreamTinyNet, the first TinyML architecture to perform multiple-frame VSA, enabling a variety of use cases that requires spatial-temporal analysis that were previously impossible to be carried out at a TinyML level. Experimental results on public-available datasets show the effectiveness and efficiency of the proposed solution. Finally, StreamTinyNet has been ported and tested on the Arduino Nicla Vision, showing the feasibility of what proposed.
Authors: Arnisha Khondaker, Nilanjan Ray
Abstract: Tuning parameters of a non-differentiable or black-box compute is challenging. Existing methods rely mostly on random sampling or grid sampling from the parameter space. Further, with all the current methods, it is not possible to supply any input specific parameters to the black-box. To the best of our knowledge, for the first time, we are able to learn input-specific parameters for a black box in this work. As a test application we choose a popular image denoising method BM3D as our black-box compute. Then, we use a differentiable surrogate model (a neural network) to approximate the black-box behaviour. Next, another neural network is used in an end-to-end fashion to learn input instance-specific parameters for the black-box. Drawing inspiration from the work of Tseng et al. [1] , we applied our method to the Smartphone Image Denoising Dataset (SIDD) for image denoising. The results are compelling, demonstrating a significant increase in PSNR and a notable improvement in SSIM nearing 0.93. Experimental results underscore the effectiveness of our approach in achieving substantial improvements in both model performance and optimization efficiency. For code and implementation details, please refer to our GitHub repository. [1] Ethan Tseng, Felix Yu, Yuting Yang, Fahim Mannan, Karl St. Arnaud, Derek Nowrouzezahrai, Jean-Francois Lalonde, and Felix Heide. Hyperparameter optimization in black-box image processing using differentiable proxies. ACM Transactions on Graphics (TOG), 38(4), 7 2019.
Authors: Changyou Chen, Han Ding, Bunyamin Sisman, Yi Xu, Ouye Xie, Benjamin Z. Yao, Son Dinh Tran, Belinda Zeng
Abstract: Diffusion-based generative modeling has been achieving state-of-the-art results on various generation tasks. Most diffusion models, however, are limited to a single-generation modeling. Can we generalize diffusion models with the ability of multi-modal generative training for more generalizable modeling? In this paper, we propose a principled way to define a diffusion model by constructing a unified multi-modal diffusion model in a common diffusion space. We define the forward diffusion process to be driven by an information aggregation from multiple types of task-data, e.g., images for a generation task and labels for a classification task. In the reverse process, we enforce information sharing by parameterizing a shared backbone denoising network with additional modality-specific decoder heads. Such a structure can simultaneously learn to generate different types of multi-modal data with a multi-task loss, which is derived from a new multi-modal variational lower bound that generalizes the standard diffusion model. We propose several multimodal generation settings to verify our framework, including image transition, masked-image training, joint image-label and joint image-representation generative modeling. Extensive experimental results on ImageNet indicate the effectiveness of our framework for various multi-modal generative modeling, which we believe is an important research direction worthy of more future explorations.
Authors: Shougao Zhang, Mengqi Zhou, Yuxi Wang, Chuanchen Luo, Rongyu Wang, Yiwei Li, Xucheng Yin, Zhaoxiang Zhang, Junran Peng
Abstract: Generating a realistic, large-scale 3D virtual city remains a complex challenge due to the involvement of numerous 3D assets, various city styles, and strict layout constraints. Existing approaches provide promising attempts at procedural content generation to create large-scale scenes using Blender agents. However, they face crucial issues such as difficulties in scaling up generation capability and achieving fine-grained control at the semantic layout level. To address these problems, we propose a novel multi-modal controllable procedural content generation method, named CityX, which enhances realistic, unbounded 3D city generation guided by multiple layout conditions, including OSM, semantic maps, and satellite images. Specifically, the proposed method contains a general protocol for integrating various PCG plugins and a multi-agent framework for transforming instructions into executable Blender actions. Through this effective framework, CityX shows the potential to build an innovative ecosystem for 3D scene generation by bridging the gap between the quality of generated assets and industrial requirements. Extensive experiments have demonstrated the effectiveness of our method in creating high-quality, diverse, and unbounded cities guided by multi-modal conditions. Our project page: https://cityx-lab.github.io.
Authors: Neha A S, Vivek Chaturvedi, Muhammad Shafique
Abstract: Vision Transformer (ViT) is becoming widely popular in automating accurate disease diagnosis in medical imaging owing to its robust self-attention mechanism. However, ViTs remain vulnerable to adversarial attacks that may thwart the diagnosis process by leading it to intentional misclassification of critical disease. In this paper, we propose a novel image classification pipeline, namely, S-E Pipeline, that performs multiple pre-processing steps that allow ViT to be trained on critical features so as to reduce the impact of input perturbations by adversaries. Our method uses a combination of segmentation and image enhancement techniques such as Contrast Limited Adaptive Histogram Equalization (CLAHE), Unsharp Masking (UM), and High-Frequency Emphasis filtering (HFE) as preprocessing steps to identify critical features that remain intact even after adversarial perturbations. The experimental study demonstrates that our novel pipeline helps in reducing the effect of adversarial attacks by 72.22% for the ViT-b32 model and 86.58% for the ViT-l32 model. Furthermore, we have shown an end-to-end deployment of our proposed method on the NVIDIA Jetson Orin Nano board to demonstrate its practical use case in modern hand-held devices that are usually resource-constrained.
Authors: Tim R\"adsch, Annika Reinke, Vivienn Weru, Minu D. Tizabi, Nicholas Heller, Fabian Isensee, Annette Kopp-Schneider, Lena Maier-Hein
Abstract: This paper does not describe a novel method. Instead, it studies an essential foundation for reliable benchmarking and ultimately real-world application of AI-based image analysis: generating high-quality reference annotations. Previous research has focused on crowdsourcing as a means of outsourcing annotations. However, little attention has so far been given to annotation companies, specifically regarding their internal quality assurance (QA) processes. Therefore, our aim is to evaluate the influence of QA employed by annotation companies on annotation quality and devise methodologies for maximizing data annotation efficacy. Based on a total of 57,648 instance segmented images obtained from a total of 924 annotators and 34 QA workers from four annotation companies and Amazon Mechanical Turk (MTurk), we derived the following insights: (1) Annotation companies perform better both in terms of quantity and quality compared to the widely used platform MTurk. (2) Annotation companies' internal QA only provides marginal improvements, if any. However, improving labeling instructions instead of investing in QA can substantially boost annotation performance. (3) The benefit of internal QA depends on specific image characteristics. Our work could enable researchers to derive substantially more value from a fixed annotation budget and change the way annotation companies conduct internal QA.
Authors: Muhammad Alberb, Marawan Elbatel, Aya Elgebaly, Ricardo Montoya-del-Angel, Xiaomeng Li, Robert Mart\'i
Abstract: Digital Breast Tomosynthesis (DBT) is an advanced breast imaging modality that offers superior lesion detection accuracy compared to conventional mammography, albeit at the trade-off of longer reading time. Accelerating lesion detection from DBT using deep learning is hindered by limited data availability and huge annotation costs. A possible solution to this issue could be to leverage the information provided by a more widely available modality, such as mammography, to enhance DBT lesion detection. In this paper, we present a novel framework, CoMoTo, for improving lesion detection in DBT. Our framework leverages unpaired mammography data to enhance the training of a DBT model, improving practicality by eliminating the need for mammography during inference. Specifically, we propose two novel components, Lesion-specific Knowledge Distillation (LsKD) and Intra-modal Point Alignment (ImPA). LsKD selectively distills lesion features from a mammography teacher model to a DBT student model, disregarding background features. ImPA further enriches LsKD by ensuring the alignment of lesion features within the teacher before distilling knowledge to the student. Our comprehensive evaluation shows that CoMoTo is superior to traditional pretraining and image-level KD, improving performance by 7% Mean Sensitivity under low-data setting. Our code is available at https://github.com/Muhammad-Al-Barbary/CoMoTo .
Authors: Hasib Zunair, A. Ben Hamza
Abstract: Localizing objects in an unsupervised manner poses significant challenges due to the absence of key visual information such as the appearance, type and number of objects, as well as the lack of labeled object classes typically available in supervised settings. While recent approaches to unsupervised object localization have demonstrated significant progress by leveraging self-supervised visual representations, they often require computationally intensive training processes, resulting in high resource demands in terms of computation, learnable parameters, and data. They also lack explicit modeling of visual context, potentially limiting their accuracy in object localization. To tackle these challenges, we propose a single-stage learning framework, dubbed PEEKABOO, for unsupervised object localization by learning context-based representations at both the pixel- and shape-level of the localized objects through image masking. The key idea is to selectively hide parts of an image and leverage the remaining image information to infer the location of objects without explicit supervision. The experimental results, both quantitative and qualitative, across various benchmark datasets, demonstrate the simplicity, effectiveness and competitive performance of our approach compared to state-of-the-art methods in both single object discovery and unsupervised salient object detection tasks. Code and pre-trained models are available at: https://github.com/hasibzunair/peekaboo
Authors: Xin Zhang, Yuqi Song, Wyatt McCurdy, Xiaofeng Wang, Fei Zuo
Abstract: Convolutional neural networks (CNNs) have gained increasing popularity and versatility in recent decades, finding applications in diverse domains. These remarkable achievements are greatly attributed to the support of extensive datasets with precise labels. However, annotating image datasets is intricate and complex, particularly in the case of multi-label datasets. Hence, the concept of partial-label setting has been proposed to reduce annotation costs, and numerous corresponding solutions have been introduced. The evaluation methods for these existing solutions have been primarily based on accuracy. That is, their performance is assessed by their predictive accuracy on the test set. However, we insist that such an evaluation is insufficient and one-sided. On one hand, since the quality of the test set has not been evaluated, the assessment results are unreliable. On the other hand, the partial-label problem may also be raised by undergoing adversarial attacks. Therefore, incorporating robustness into the evaluation system is crucial. For this purpose, we first propose two attack models to generate multiple partial-label datasets with varying degrees of label missing rates. Subsequently, we introduce a lightweight partial-label solution using pseudo-labeling techniques and a designed loss function. Then, we employ D-Score to analyze both the proposed and existing methods to determine whether they can enhance robustness while improving accuracy. Extensive experimental results demonstrate that while certain methods may improve accuracy, the enhancement in robustness is not significant, and in some cases, it even diminishes.
Authors: Binay Kumar Singh, Niels Da Vitoria Lobo
Abstract: With the growing advances in deep learning based technologies the detection and identification of co-occurring objects is a challenging task which has many applications in areas such as, security and surveillance. In this paper, we propose a novel framework called SDLNet- Statistical analysis with Deep Learning Network that identifies co-occurring objects in conjunction with base objects in multilabel object categories. The pipeline of proposed work is implemented in two stages: in the first stage of SDLNet we deal with multilabel detectors for discovering labels, and in the second stage we perform co-occurrence matrix analysis. In co-occurrence matrix analysis, we learn co-occurrence statistics by setting base classes and frequently occurring classes, following this we build association rules and generate frequent patterns. The crucial part of SDLNet is recognizing base classes and making consideration for co-occurring classes. Finally, the generated co-occurrence matrix based on frequent patterns will show base classes and their corresponding co-occurring classes. SDLNet is evaluated on two publicly available datasets: Pascal VOC and MS-COCO. The experimental results on these benchmark datasets are reported in Sec 4.
Authors: Qing Su, Shihao Ji
Abstract: Distillation-based self-supervised learning typically leads to more compressed representations due to its radical clustering process and the implementation of a sharper target distribution. To overcome this limitation and preserve more information from input, we introduce UDI, conceptualized as Unsqueezed Distillation-based self-supervised learning (SSL). UDI enriches the learned representation by encouraging multimodal prediction distilled from a consolidated profile of local predictions that are derived via stratified sampling. Our evaluations show that UDI not only promotes semantically meaningful representations at instance level, delivering superior or competitive results to state-of-the-art SSL methods in image classification, but also effectively preserves the nuisance of input, which yields significant improvement in dense prediction tasks, including object detection and segmentation. Additionally, UDI performs competitively in low-shot image classification, improving the scalability of joint-embedding pipelines. Various visualizations and ablation studies are presented to further elucidate the mechanisms behind UDI. Our source code is available at https://github.com/ISL-CV/udi.
Authors: Thomas Manzini, Priyankari Perali, Raisa Karnik, Robin Murphy
Abstract: This document presents the Center for Robot Assisted Search And Rescue - Uncrewed Aerial Systems - Disaster Response Overhead Inspection Dataset (CRASAR-U-DROIDs) for building damage assessment and spatial alignment collected from small uncrewed aerial systems (sUAS) geospatial imagery. This dataset is motivated by the increasing use of sUAS in disaster response and the lack of previous work in utilizing high-resolution geospatial sUAS imagery for machine learning and computer vision models, the lack of alignment with operational use cases, and with hopes of enabling further investigations between sUAS and satellite imagery. The CRASAR-U-DRIODs dataset consists of fifty-two (52) orthomosaics from ten (10) federally declared disasters (Hurricane Ian, Hurricane Ida, Hurricane Harvey, Hurricane Idalia, Hurricane Laura, Hurricane Michael, Musset Bayou Fire, Mayfield Tornado, Kilauea Eruption, and Champlain Towers Collapse) spanning 67.98 square kilometers (26.245 square miles), containing 21,716 building polygons and damage labels, and 7,880 adjustment annotations. The imagery was tiled and presented in conjunction with overlaid building polygons to a pool of 130 annotators who provided human judgments of damage according to the Joint Damage Scale. These annotations were then reviewed via a two-stage review process in which building polygon damage labels were first reviewed individually and then again by committee. Additionally, the building polygons have been aligned spatially to precisely overlap with the imagery to enable more performant machine learning models to be trained. It appears that CRASAR-U-DRIODs is the largest labeled dataset of sUAS orthomosaic imagery.
Authors: Heng Fang, Sheng Huang, Wenhao Tang, Luwen Huangfu, Bo Liu
Abstract: Multiple Instance Learning (MIL) represents the predominant framework in Whole Slide Image (WSI) classification, covering aspects such as sub-typing, diagnosis, and beyond. Current MIL models predominantly rely on instance-level features derived from pretrained models such as ResNet. These models segment each WSI into independent patches and extract features from these local patches, leading to a significant loss of global spatial context and restricting the model's focus to merely local features. To address this issue, we propose a novel MIL framework, named SAM-MIL, that emphasizes spatial contextual awareness and explicitly incorporates spatial context by extracting comprehensive, image-level information. The Segment Anything Model (SAM) represents a pioneering visual segmentation foundational model that can capture segmentation features without the need for additional fine-tuning, rendering it an outstanding tool for extracting spatial context directly from raw WSIs. Our approach includes the design of group feature extraction based on spatial context and a SAM-Guided Group Masking strategy to mitigate class imbalance issues. We implement a dynamic mask ratio for different segmentation categories and supplement these with representative group features of categories. Moreover, SAM-MIL divides instances to generate additional pseudo-bags, thereby augmenting the training set, and introduces consistency of spatial context across pseudo-bags to further enhance the model's performance. Experimental results on the CAMELYON-16 and TCGA Lung Cancer datasets demonstrate that our proposed SAM-MIL model outperforms existing mainstream methods in WSIs classification. Our open-source implementation code is is available at https://github.com/FangHeng/SAM-MIL.
Authors: Shichen Qu, Xian Tao, Zhen Qu, Xinyi Gong, Zhengtao Zhang, Mukesh Prasad
Abstract: Unsupervised anomaly localization on industrial textured images has achieved remarkable results through reconstruction-based methods, yet existing approaches based on image reconstruction and feature reconstruc-tion each have their own shortcomings. Firstly, image-based methods tend to reconstruct both normal and anomalous regions well, which lead to over-generalization. Feature-based methods contain a large amount of distin-guishable semantic information, however, its feature structure is redundant and lacks anomalous information, which leads to significant reconstruction errors. In this paper, we propose an Anomaly Localization method based on Mamba with Feature Reconstruction and Refinement(ALMRR) which re-constructs semantic features based on Mamba and then refines them through a feature refinement module. To equip the model with prior knowledge of anomalies, we enhance it by adding artificially simulated anomalies to the original images. Unlike image reconstruction or repair, the features of synthesized defects are repaired along with those of normal areas. Finally, the aligned features containing rich semantic information are fed in-to the refinement module to obtain the anomaly map. Extensive experiments have been conducted on the MVTec-AD-Textured dataset and other real-world industrial dataset, which has demonstrated superior performance com-pared to state-of-the-art (SOTA) methods.
Authors: Haoran Zhu, Yifan Zhou, Chang Xu, Ruixiang Zhang, Wen Yang
Abstract: Fine-Grained Object Detection (FGOD) is a critical task in high-resolution aerial image analysis. This letter introduces Orthogonal Mapping (OM), a simple yet effective method aimed at addressing the challenge of semantic confusion inherent in FGOD. OM introduces orthogonal constraints in the feature space by decoupling features from the last layer of the classification branch with a class-wise orthogonal vector basis. This effectively mitigates semantic confusion and enhances classification accuracy. Moreover, OM can be seamlessly integrated into mainstream object detectors. Extensive experiments conducted on three FGOD datasets (FAIR1M, ShipRSImageNet, and MAR20) demonstrate the effectiveness and superiority of the proposed approach. Notably, with just one line of code, OM achieves a 4.08% improvement in mean Average Precision (mAP) over FCOS on the ShipRSImageNet dataset. Codes are released at https://github.com/ZhuHaoranEIS/Orthogonal-FGOD.
Authors: Bo Li
Abstract: This paper study one challenging issue in incomplete multi-view clustering, where valuable complementary information from other views is always ignored. To be specific, we propose a framework that effectively balances Complementarity and Consistency information in Incomplete Multi-view Clustering (CoCo-IMC). Specifically, we design a dual network of delayed activation, which achieves a balance of complementarity and consistency among different views. The delayed activation could enriches the complementarity information that was ignored during consistency learning. Then, we recover the incomplete information and enhance the consistency learning by minimizing the conditional entropy and maximizing the mutual information across different views. This could be the first theoretical attempt to incorporate delayed activation into incomplete data recovery and the balance of complementarity and consistency. We have proved the effectiveness of CoCo-IMC in extensive comparative experiments with 12 state-of-the-art baselines on four publicly available datasets.
Authors: Saideep Kilaru, Kothamasu Jayachandra, Tanishka Yagneshwar, Suchi Kumari
Abstract: In recent years, the focus is on improving the diagnosis of diabetic retinopathy (DR) using machine learning and deep learning technologies. Researchers have explored various approaches, including the use of high-definition medical imaging, AI-driven algorithms such as convolutional neural networks (CNNs) and generative adversarial networks (GANs). Among all the available tools, CNNs have emerged as a preferred tool due to their superior classification accuracy and efficiency. Although the accuracy of CNNs is comparatively better but it can be improved by introducing some hybrid models by combining various machine learning and deep learning models. Therefore, in this paper, an ensemble learning technique is proposed for early detection and management of DR with higher accuracy. The proposed model is tested on the APTOS dataset and it is showing supremacy on the validation accuracy ($99\%)$ in comparison to the previous models. Hence, the model can be helpful for early detection and treatment of the DR, thereby enhancing the overall quality of care for affected individuals.
Authors: Haicheng Liao, Haoyu Sun, Huanming Shen, Chengyue Wang, Kahou Tam, Chunlin Tian, Li Li, Chengzhong Xu, Zhenning Li
Abstract: Accurately and promptly predicting accidents among surrounding traffic agents from camera footage is crucial for the safety of autonomous vehicles (AVs). This task presents substantial challenges stemming from the unpredictable nature of traffic accidents, their long-tail distribution, the intricacies of traffic scene dynamics, and the inherently constrained field of vision of onboard cameras. To address these challenges, this study introduces a novel accident anticipation framework for AVs, termed CRASH. It seamlessly integrates five components: object detector, feature extractor, object-aware module, context-aware module, and multi-layer fusion. Specifically, we develop the object-aware module to prioritize high-risk objects in complex and ambiguous environments by calculating the spatial-temporal relationships between traffic agents. In parallel, the context-aware is also devised to extend global visual information from the temporal to the frequency domain using the Fast Fourier Transform (FFT) and capture fine-grained visual features of potential objects and broader context cues within traffic scenes. To capture a wider range of visual cues, we further propose a multi-layer fusion that dynamically computes the temporal dependencies between different scenes and iteratively updates the correlations between different visual features for accurate and timely accident prediction. Evaluated on real-world datasets--Dashcam Accident Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D) datasets--our model surpasses existing top baselines in critical evaluation metrics like Average Precision (AP) and mean Time-To-Accident (mTTA). Importantly, its robustness and adaptability are particularly evident in challenging driving scenarios with missing or limited training data, demonstrating significant potential for application in real-world autonomous driving systems.
Authors: Yudara Kularathne, Prathapa Janitha, Sithira Ambepitiya, Prarththanan Sothyrajah, Thanveer Ahamed, Dinuka Wijesundara
Abstract: Rapid development of disease detection models using computer vision is crucial in responding to medical emergencies, such as epidemics or bioterrorism events. Traditional data collection methods are often too slow in these scenarios, requiring innovative approaches for quick, reliable model generation from minimal data. Our study introduces a novel approach by constructing a comprehensive computer vision model to detect Mpox lesions using only synthetic data. Initially, these models generated a diverse set of synthetic images representing Mpox lesions on various body parts (face, back, chest, leg, neck, arm) across different skin tones as defined by the Fitzpatrick scale (fair, brown, dark skin). Subsequently, we trained and tested a vision model with this synthetic dataset to evaluate the diffusion models' efficacy in producing high-quality training data and its impact on the vision model's medical image recognition performance. The results were promising; the vision model achieved a 97% accuracy rate, with 96% precision and recall for Mpox cases, and similarly high metrics for normal and other skin disorder cases, demonstrating its ability to correctly identify true positives and minimize false positives. The model achieved an F1-Score of 96% for Mpox cases and 98% for normal and other skin disorders, reflecting a balanced precision-recall relationship, thus ensuring reliability and robustness in its predictions. Our proposed SynthVision methodology indicates the potential to develop accurate computer vision models with minimal data input for future medical emergencies.
Authors: Rita Frieske, Bertrand E. Shi
Abstract: ERIT is a novel multimodal dataset designed to facilitate research in a lightweight multimodal fusion. It contains text and image data collected from videos of elderly individuals reacting to various situations, as well as seven emotion labels for each data sample. Because of the use of labeled images of elderly users reacting emotionally, it is also facilitating research on emotion recognition in an underrepresented age group in machine learning visual emotion recognition. The dataset is validated through comprehensive experiments indicating its importance in neural multimodal fusion research.
Authors: Eunice Yiu, Maan Qraitem, Charlie Wong, Anisa Noor Majhi, Yutong Bai, Shiry Ginosar, Alison Gopnik, Kate Saenko
Abstract: This paper investigates visual analogical reasoning in large multimodal models (LMMs) compared to human adults and children. A "visual analogy" is an abstract rule inferred from one image and applied to another. While benchmarks exist for testing visual reasoning in LMMs, they require advanced skills and omit basic visual analogies that even young children can make. Inspired by developmental psychology, we propose a new benchmark of 1,400 visual transformations of everyday objects to test LMMs on visual analogical reasoning and compare them to children and adults. We structure the evaluation into three stages: identifying what changed (e.g., color, number, etc.), how it changed (e.g., added one object), and applying the rule to new scenarios. Our findings show that while models like GPT-4V, LLaVA-1.5, and MANTIS identify the "what" effectively, they struggle with quantifying the "how" and extrapolating this rule to new objects. In contrast, children and adults exhibit much stronger analogical reasoning at all three stages. Additionally, the strongest tested model, GPT-4V, performs better in tasks involving simple visual attributes like color and size, correlating with quicker human adult response times. Conversely, more complex tasks such as number, rotation, and reflection, which necessitate extensive cognitive processing and understanding of the 3D physical world, present more significant challenges. Altogether, these findings highlight the limitations of training models on data that primarily consists of 2D images and text.
Authors: Chaofan Gan, Yuanpeng Tu, Yuxi Li, Weiyao Lin
Abstract: With the recent burst of 2D and 3D data, cross-modal retrieval has attracted increasing attention recently. However, manual labeling by non-experts will inevitably introduce corrupted annotations given ambiguous 2D/3D content. Though previous works have addressed this issue by designing a naive division strategy with hand-crafted thresholds, their performance generally exhibits great sensitivity to the threshold value. Besides, they fail to fully utilize the valuable supervisory signals within each divided subset. To tackle this problem, we propose a Divide-and-conquer 2D-3D cross-modal Alignment and Correction framework (DAC), which comprises Multimodal Dynamic Division (MDD) and Adaptive Alignment and Correction (AAC). Specifically, the former performs accurate sample division by adaptive credibility modeling for each sample based on the compensation information within multimodal loss distribution. Then in AAC, samples in distinct subsets are exploited with different alignment strategies to fully enhance the semantic compactness and meanwhile alleviate over-fitting to noisy labels, where a self-correction strategy is introduced to improve the quality of representation. Moreover. To evaluate the effectiveness in real-world scenarios, we introduce a challenging noisy benchmark, namely Objaverse-N200, which comprises 200k-level samples annotated with 1156 realistic noisy labels. Extensive experiments on both traditional and the newly proposed benchmarks demonstrate the generality and superiority of our DAC, where DAC outperforms state-of-the-art models by a large margin. (i.e., with +5.9% gain on ModelNet40 and +5.8% on Objaverse-N200).
Authors: Jen Hong Tan
Abstract: In this paper, we explore a strategy that uses Mixture-of-Experts (MoE) to streamline, rather than augment, vision transformers. Each expert in an MoE layer is a SwiGLU feedforward network, where V and W2 are shared across the layer. No complex attention or convolutional mechanisms are employed. Depth-wise scaling is applied to progressively reduce the size of the hidden layer and the number of experts is increased in stages. Grouped query attention is used. We studied the proposed approach with and without pre-training on small datasets and investigated whether transfer learning works at this scale. We found that the architecture is competitive even at a size of 0.67M parameters.
Authors: Chia-Chia Chen, Chi-Han Peng
Abstract: We present a novel discrete optimization-based approach to generate downsampled versions of binary images that are guaranteed to have the same topology as the original, measured by the zeroth and first Betti numbers of the black regions, while having good similarity to the original image as measured by IoU and Dice scores. To our best knowledge, all existing binary image downsampling methods do not have such topology-preserving guarantees. We also implemented a baseline morphological operation (dilation)-based approach that always generates topologically correct results. However, we found the similarity scores to be much worse. We demonstrate several applications of our approach. First, generating smaller versions of medical image segmentation masks for easier human inspection. Second, improving the efficiency of binary image operations, including persistent homology computation and shortest path computation, by substituting the original images with smaller ones. In particular, the latter is a novel application that is made feasible only by the full topology-preservation guarantee of our method.
Authors: Shuming Liu, Lin Sui, Chen-Lin Zhang, Fangzhou Mu, Chen Zhao, Bernard Ghanem
Abstract: As a fundamental task in long-form video understanding, temporal action detection (TAD) aims to capture inherent temporal relations in untrimmed videos and identify candidate actions with precise boundaries. Over the years, various networks, including convolutions, graphs, and transformers, have been explored for effective temporal modeling for TAD. However, these modules typically treat past and future information equally, overlooking the crucial fact that changes in action boundaries are essentially causal events. Inspired by this insight, we propose leveraging the temporal causality of actions to enhance TAD representation by restricting the model's access to only past or future context. We introduce CausalTAD, which combines causal attention and causal Mamba to achieve state-of-the-art performance on multiple benchmarks. Notably, with CausalTAD, we ranked 1st in the Action Recognition, Action Detection, and Audio-Based Interaction Detection tracks at the EPIC-Kitchens Challenge 2024, as well as 1st in the Moment Queries track at the Ego4D Challenge 2024. Our code is available at https://github.com/sming256/OpenTAD/causaltad.
Authors: Haonan Zheng, Xinyang Deng, Wen Jiang, Wenrui Li
Abstract: With Vision-Language Pre-training (VLP) models demonstrating powerful multimodal interaction capabilities, the application scenarios of neural networks are no longer confined to unimodal domains but have expanded to more complex multimodal V+L downstream tasks. The security vulnerabilities of unimodal models have been extensively examined, whereas those of VLP models remain challenging. We note that in CV models, the understanding of images comes from annotated information, while VLP models are designed to learn image representations directly from raw text. Motivated by this discrepancy, we developed the Feature Guidance Attack (FGA), a novel method that uses text representations to direct the perturbation of clean images, resulting in the generation of adversarial images. FGA is orthogonal to many advanced attack strategies in the unimodal domain, facilitating the direct application of rich research findings from the unimodal to the multimodal scenario. By appropriately introducing text attack into FGA, we construct Feature Guidance with Text Attack (FGA-T). Through the interaction of attacking two modalities, FGA-T achieves superior attack effects against VLP models. Moreover, incorporating data augmentation and momentum mechanisms significantly improves the black-box transferability of FGA-T. Our method demonstrates stable and effective attack capabilities across various datasets, downstream tasks, and both black-box and white-box settings, offering a unified baseline for exploring the robustness of VLP models.
Authors: Vedanshu, MM Tripathi, Bhavnesh Jaint
Abstract: The integration of large language models (LLMs) with vision-language (VL) tasks has been a transformative development in the realm of artificial intelligence, highlighting the potential of LLMs as a versatile general-purpose chatbot. However, the current trend in this evolution focuses on the integration of vision and language to create models that can operate in more diverse and real-world contexts. We present a novel approach, termed Bottleneck Adapter, specifically crafted for enhancing the multimodal functionalities of these complex models, enabling joint optimization of the entire multimodal LLM framework through a process known as Multimodal Model Tuning (MMT). Our approach utilizes lightweight adapters to connect the image encoder and LLM without the need for large, complex neural networks. Unlike the conventional modular training schemes, our approach adopts an end-to-end optimization regime, which, when combined with the adapters, facilitates the joint optimization using a significantly smaller parameter set. Our method exhibits robust performance with 90.12\% accuracy, outperforming both human-level performance (88.4\%) and LaVIN-7B (89.41\%).
Authors: Yifan Li, Yikai Wang, Yanwei Fu, Dongyu Ru, Zheng Zhang, Tong He
Abstract: Visual-Language Alignment (VLA) has gained a lot of attention since CLIP's groundbreaking work. Although CLIP performs well, the typical direct latent feature alignment lacks clarity in its representation and similarity scores. On the other hand, lexical representation, a vector whose element represents the similarity between the sample and a word from the vocabulary, is a natural sparse representation and interpretable, providing exact matches for individual words. However, lexical representations is difficult to learn due to no ground-truth supervision and false-discovery issues, and thus requires complex design to train effectively. In this paper, we introduce LexVLA, a more interpretable VLA framework by learning a unified lexical representation for both modalities without complex design. We use DINOv2 as our visual model for its local-inclined features and Llama 2, a generative language model, to leverage its in-context lexical prediction ability. To avoid the false discovery, we propose an overuse penalty to refrain the lexical representation from falsely frequently activating meaningless words. We demonstrate that these two pre-trained uni-modal models can be well-aligned by fine-tuning on modest multi-modal dataset and avoid intricate training configurations. On cross-modal retrieval benchmarks, LexVLA, trained on the CC-12M multi-modal dataset, outperforms baselines fine-tuned on larger datasets (e.g., YFCC15M) and those trained from scratch on even bigger datasets (e.g., 1.1B data, including CC-12M). We conduct extensive experiments to analyze LexVLA.
Authors: Pablo Hern\'andez-C\'amara, Jorge Vila-Tom\'as, Paula Dauden-Oliver, Nuria Alabau-Bosque, Valero Laparra, Jes\'us Malo
Abstract: Autonomous driving is a challenging scenario for image segmentation due to the presence of uncontrolled environmental conditions and the eventually catastrophic consequences of failures. Previous work suggested that a biologically motivated computation, the so-called Divisive Normalization, could be useful to deal with image variability, but its effects have not been systematically studied over different data sources and environmental factors. Here we put segmentation U-nets augmented with Divisive Normalization to work far from training conditions to find where this adaptation is more critical. We categorize the scenes according to their radiance level and dynamic range (day/night), and according to their achromatic/chromatic contrasts. We also consider video game (synthetic) images to broaden the range of environments. We check the performance in the extreme percentiles of such categorization. Then, we push the limits further by artificially modifying the images in perceptually/environmentally relevant dimensions: luminance, contrasts and spectral radiance. Results show that neural networks with Divisive Normalization get better results in all the scenarios and their performance remains more stable with regard to the considered environmental factors and nature of the source. Finally, we explain the improvements in segmentation performance in two ways: (1) by quantifying the invariance of the responses that incorporate Divisive Normalization, and (2) by illustrating the adaptive nonlinearity of the different layers that depends on the local activity.
Authors: Zhicheng Cai, Hao Zhu, Qiu Shen, Xinran Wang, Xun Cao
Abstract: Representing signals using coordinate networks dominates the area of inverse problems recently, and is widely applied in various scientific computing tasks. Still, there exists an issue of spectral bias in coordinate networks, limiting the capacity to learn high-frequency components. This problem is caused by the pathological distribution of the neural tangent kernel's (NTK's) eigenvalues of coordinate networks. We find that, this pathological distribution could be improved using classical normalization techniques (batch normalization and layer normalization), which are commonly used in convolutional neural networks but rarely used in coordinate networks. We prove that normalization techniques greatly reduces the maximum and variance of NTK's eigenvalues while slightly modifies the mean value, considering the max eigenvalue is much larger than the most, this variance change results in a shift of eigenvalues' distribution from a lower one to a higher one, therefore the spectral bias could be alleviated. Furthermore, we propose two new normalization techniques by combining these two techniques in different ways. The efficacy of these normalization techniques is substantiated by the significant improvements and new state-of-the-arts achieved by applying normalization-based coordinate networks to various tasks, including the image compression, computed tomography reconstruction, shape representation, magnetic resonance imaging, novel view synthesis and multi-view stereo reconstruction.
Authors: Jian Wang, Jing Wang, Shenghui Rong, Bo He
Abstract: Underwater monocular depth estimation serves as the foundation for tasks such as 3D reconstruction of underwater scenes. However, due to the influence of light and medium, the underwater environment undergoes a distinctive imaging process, which presents challenges in accurately estimating depth from a single image. The existing methods fail to consider the unique characteristics of underwater environments, leading to inadequate estimation results and limited generalization performance. Furthermore, underwater depth estimation requires extracting and fusing both local and global features, which is not fully explored in existing methods. In this paper, an end-to-end learning framework for underwater monocular depth estimation called UMono is presented, which incorporates underwater image formation model characteristics into network architecture, and effectively utilize both local and global features of underwater image. Experimental results demonstrate that the proposed method is effective for underwater monocular depth estimation and outperforms the existing methods in both quantitative and qualitative analyses.
Authors: Gayoon Choi, Taejin Jeong, Sujung Hong, Jaehoon Joo, Seong Jae Hwang
Abstract: Point-based image editing enables accurate and flexible control through content dragging. However, the role of text embedding in the editing process has not been thoroughly investigated. A significant aspect that remains unexplored is the interaction between text and image embeddings. In this study, we show that during the progressive editing of an input image in a diffusion model, the text embedding remains constant. As the image embedding increasingly diverges from its initial state, the discrepancy between the image and text embeddings presents a significant challenge. Moreover, we found that the text prompt significantly influences the dragging process, particularly in maintaining content integrity and achieving the desired manipulation. To utilize these insights, we propose DragText, which optimizes text embedding in conjunction with the dragging process to pair with the modified image embedding. Simultaneously, we regularize the text optimization process to preserve the integrity of the original text prompt. Our approach can be seamlessly integrated with existing diffusion-based drag methods with only a few lines of code.
Authors: Pengfei Jiang, Mingbao Lin, Fei Chao, Rongrong Ji
Abstract: Current methods commonly utilize three-branch structures of inversion, reconstruction, and editing, to tackle consistent image editing task. However, these methods lack control over the generation position of the edited object and have issues with background preservation. To overcome these limitations, we propose a tuning-free method with only two branches: inversion and editing. This approach allows users to simultaneously edit the object's action and control the generation position of the edited object. Additionally, it achieves improved background preservation. Specifically, we transfer the edited object information to the target area and repair or preserve the background of other areas during the inversion process at a specific time step. In the editing stage, we use the image features in self-attention to query the key and value of the corresponding time step in the inversion to achieve consistent image editing. Impressive image editing results and quantitative evaluation demonstrate the effectiveness of our method. The code is available at https://github.com/mobiushy/move-act.
Authors: Gwanhyeong Koo, Sunjae Yoon, Ji Woo Hong, Chang D. Yoo
Abstract: Current image editing methods primarily utilize DDIM Inversion, employing a two-branch diffusion approach to preserve the attributes and layout of the original image. However, these methods encounter challenges with non-rigid edits, which involve altering the image's layout or structure. Our comprehensive analysis reveals that the high-frequency components of DDIM latent, crucial for retaining the original image's key features and layout, significantly contribute to these limitations. Addressing this, we introduce FlexiEdit, which enhances fidelity to input text prompts by refining DDIM latent, by reducing high-frequency components in targeted editing areas. FlexiEdit comprises two key components: (1) Latent Refinement, which modifies DDIM latent to better accommodate layout adjustments, and (2) Edit Fidelity Enhancement via Re-inversion, aimed at ensuring the edits more accurately reflect the input text prompts. Our approach represents notable progress in image editing, particularly in performing complex non-rigid edits, showcasing its enhanced capability through comparative experiments.
Authors: Sukwon Yun, Jie Peng, Alexandro E. Trevino, Chanyoung Park, Tianlong Chen
Abstract: Recent advancements in graph-based approaches for multiplexed immunofluorescence (mIF) images have significantly propelled the field forward, offering deeper insights into patient-level phenotyping. However, current graph-based methodologies encounter two primary challenges: (1) Cellular Heterogeneity, where existing approaches fail to adequately address the inductive biases inherent in graphs, particularly the homophily characteristic observed in cellular connectivity and; (2) Scalability, where handling cellular graphs from high-dimensional images faces difficulties in managing a high number of cells. To overcome these limitations, we introduce Mew, a novel framework designed to efficiently process mIF images through the lens of multiplex network. Mew innovatively constructs a multiplex network comprising two distinct layers: a Voronoi network for geometric information and a Cell-type network for capturing cell-wise homogeneity. This framework equips a scalable and efficient Graph Neural Network (GNN), capable of processing the entire graph during training. Furthermore, Mew integrates an interpretable attention module that autonomously identifies relevant layers for image classification. Extensive experiments on a real-world patient dataset from various institutions highlight Mew's remarkable efficacy and efficiency, marking a significant advancement in mIF image analysis. The source code of Mew can be found here: \url{https://github.com/UNITES-Lab/Mew}
Authors: Shahab Saquib Sohail, Yassine Himeur, Hamza Kheddar, Abbes Amira, Fodil Fadli, Shadi Atalla, Abigail Copiaco, Wathiq Mansoor
Abstract: The 3D point cloud (3DPC) has significantly evolved and benefited from the advance of deep learning (DL). However, the latter faces various issues, including the lack of data or annotated data, the existence of a significant gap between training data and test data, and the requirement for high computational resources. To that end, deep transfer learning (DTL), which decreases dependency and costs by utilizing knowledge gained from a source data/task in training a target data/task, has been widely investigated. Numerous DTL frameworks have been suggested for aligning point clouds obtained from several scans of the same scene. Additionally, DA, which is a subset of DTL, has been modified to enhance the point cloud data's quality by dealing with noise and missing points. Ultimately, fine-tuning and DA approaches have demonstrated their effectiveness in addressing the distinct difficulties inherent in point cloud data. This paper presents the first review shedding light on this aspect. it provides a comprehensive overview of the latest techniques for understanding 3DPC using DTL and domain adaptation (DA). Accordingly, DTL's background is first presented along with the datasets and evaluation metrics. A well-defined taxonomy is introduced, and detailed comparisons are presented, considering different aspects such as different knowledge transfer strategies, and performance. The paper covers various applications, such as 3DPC object detection, semantic labeling, segmentation, classification, registration, downsampling/upsampling, and denoising. Furthermore, the article discusses the advantages and limitations of the presented frameworks, identifies open challenges, and suggests potential research directions.
Authors: Tim J. M. Jaspers, Ronald L. P. D. de Jong, Yasmina Al Khalil, Tijn Zeelenberg, Carolus H. J. Kusters, Yiping Li, Romy C. van Jaarsveld, Franciscus H. A. Bakker, Jelle P. Ruurda, Willem M. Brinkman, Peter H. N. De With, Fons van der Sommen
Abstract: Over the past decade, computer vision applications in minimally invasive surgery have rapidly increased. Despite this growth, the impact of surgical computer vision remains limited compared to other medical fields like pathology and radiology, primarily due to the scarcity of representative annotated data. Whereas transfer learning from large annotated datasets such as ImageNet has been conventionally the norm to achieve high-performing models, recent advancements in self-supervised learning (SSL) have demonstrated superior performance. In medical image analysis, in-domain SSL pretraining has already been shown to outperform ImageNet-based initialization. Although unlabeled data in the field of surgical computer vision is abundant, the diversity within this data is limited. This study investigates the role of dataset diversity in SSL for surgical computer vision, comparing procedure-specific datasets against a more heterogeneous general surgical dataset across three different downstream surgical applications. The obtained results show that using solely procedure-specific data can lead to substantial improvements of 13.8%, 9.5%, and 36.8% compared to ImageNet pretraining. However, extending this data with more heterogeneous surgical data further increases performance by an additional 5.0%, 5.2%, and 2.5%, suggesting that increasing diversity within SSL data is beneficial for model performance. The code and pretrained model weights are made publicly available at https://github.com/TimJaspers0801/SurgeNet.
Authors: Zhiheng Li, Yubo Cui, Jiexi Zhong, Zheng Fang
Abstract: Moving object segmentation based on LiDAR is a crucial and challenging task for autonomous driving and mobile robotics. Most approaches explore spatio-temporal information from LiDAR sequences to predict moving objects in the current frame. However, they often focus on transferring temporal cues in a single inference and regard every prediction as independent of others. This may cause inconsistent segmentation results for the same object in different frames. To overcome this issue, we propose a streaming network with a memory mechanism, called StreamMOS, to build the association of features and predictions among multiple inferences. Specifically, we utilize a short-term memory to convey historical features, which can be regarded as spatial prior of moving objects and adopted to enhance current inference by temporal fusion. Meanwhile, we build a long-term memory to store previous predictions and exploit them to refine the present forecast at voxel and instance levels through voting. Besides, we present multi-view encoder with cascade projection and asymmetric convolution to extract motion feature of objects in different representations. Extensive experiments validate that our algorithm gets competitive performance on SemanticKITTI and Sipailou Campus datasets. Code will be released at https://github.com/NEU-REAL/StreamMOS.git.
Authors: Kohei Iwano, Shogo Shibuya, Satoshi Kagiwada, Hitoshi Iyatomi
Abstract: Recently, object detection methods (OD; e.g., YOLO-based models) have been widely utilized in plant disease diagnosis. These methods demonstrate robustness to distance variations and excel at detecting small lesions compared to classification methods (CL; e.g., CNN models). However, there are issues such as low diagnostic performance for hard-to-detect diseases and high labeling costs. Additionally, since healthy cases cannot be explicitly trained, there is a risk of false positives. We propose the Hierarchical object detection and recognition framework (HODRF), a sophisticated and highly integrated two-stage system that combines the strengths of both OD and CL for plant disease diagnosis. In the first stage, HODRF uses OD to identify regions of interest (ROIs) without specifying the disease. In the second stage, CL diagnoses diseases surrounding the ROIs. HODRF offers several advantages: (1) Since OD detects only one type of ROI, HODRF can detect diseases with limited training images by leveraging its ability to identify other lesions. (2) While OD over-detects healthy cases, HODRF significantly reduces these errors by using CL in the second stage. (3) CL's accuracy improves in HODRF as it identifies diagnostic targets given as ROIs, making it less vulnerable to size changes. (4) HODRF benefits from CL's lower annotation costs, allowing it to learn from a larger number of images. We implemented HODRF using YOLOv7 for OD and EfficientNetV2 for CL and evaluated its performance on a large-scale dataset (4 crops, 20 diseased and healthy classes, 281K images). HODRF outperformed YOLOv7 alone by 5.8 to 21.5 points on healthy data and 0.6 to 7.5 points on macro F1 scores, and it improved macro F1 by 1.1 to 7.2 points over EfficientNetV2.
Authors: Abbas Mammadov, Hyungjin Chung, Jong Chul Ye
Abstract: We propose a variational inference approach to sample from the posterior distribution for solving inverse problems. From a pre-trained diffusion model, our approach trains a conditional flow model to minimize the divergence between the proposal variational distribution and the posterior distribution implicitly defined through the diffusion model. Once trained, the flow model is capable of sampling from the posterior distribution with a single NFE, amortized with respect to the measurement. The proposed method paves a new path for distilling a diffusion prior for efficient posterior sampling. We show that our method is applicable to standard signals in Euclidean space, as well as signals on manifold.
Authors: Kangil Lee, Geonuk Kim
Abstract: Vision-based inspection algorithms have significantly contributed to quality control in industrial settings, particularly in addressing structural defects like dent and contamination which are prevalent in mass production. Extensive research efforts have led to the development of related benchmarks such as MVTec AD (Bergmann et al., 2019). However, in industrial settings, there can be instances of logical defects, where acceptable items are found in unsuitable locations or product pairs do not match as expected. Recent methods tackling logical defects effectively employ knowledge distillation to generate difference maps. Knowledge distillation (KD) is used to learn normal data distribution in unsupervised manner. Despite their effectiveness, these methods often overlook the potential false negatives. Excessive similarity between the teacher network and student network can hinder the generation of a suitable difference map for logical anomaly detection. This technical report provides insights on handling potential false negatives by utilizing a simple constraint in KD-based logical anomaly detection methods. We select EfficientAD as a state-of-the-art baseline and apply a margin-based constraint to its unsupervised learning scheme. Applying this constraint, we can improve the AUROC for MVTec LOCO AD by 1.3 %.
Authors: Nuria Alabau-Bosque, Paula Daud\'en-Oliver, Jorge Vila-Tom\'as, Valero Laparra, Jes\'us Malo
Abstract: Deep architectures are the current state-of-the-art in predicting subjective image quality. Usually, these models are evaluated according to their ability to correlate with human opinion in databases with a range of distortions that may appear in digital media. However, these oversee affine transformations which may represent better the changes in the images actually happening in natural conditions. Humans can be particularly invariant to these natural transformations, as opposed to the digital ones. In this work, we evaluate state-of-the-art deep image quality metrics by assessing their invariance to affine transformations, specifically: rotation, translation, scaling, and changes in spectral illumination. We propose a methodology to assign invisibility thresholds for any perceptual metric. This methodology involves transforming the distance measured by an arbitrary metric to a common distance representation based on available subjectively rated databases. We psychophysically measure an absolute detection threshold in that common representation and express it in the physical units of each affine transform for each metric. By doing so, we allow the analyzed metrics to be directly comparable with actual human thresholds. We find that none of the state-of-the-art metrics shows human-like results under this strong test based on invisibility thresholds. This means that tuning the models exclusively to predict the visibility of generic distortions may disregard other properties of human vision as for instance invariances or invisibility thresholds.
Authors: Krishnakant Singh, Simone Schaub-Meyer, Stefan Roth
Abstract: Slot attention aims to decompose an input image into a set of meaningful object files (slots). These latent object representations enable various downstream tasks. Yet, these slots often bind to object parts, not objects themselves, especially for real-world datasets. To address this, we introduce Guided Latent Slot Diffusion - GLASS, an object-centric model that uses generated captions as a guiding signal to better align slots with objects. Our key insight is to learn the slot-attention module in the space of generated images. This allows us to repurpose the pre-trained diffusion decoder model, which reconstructs the images from the slots, as a semantic mask generator based on the generated captions. GLASS learns an object-level representation suitable for multiple tasks simultaneously, e.g., segmentation, image generation, and property prediction, outperforming previous methods. For object discovery, GLASS achieves approx. a +35% and +10% relative improvement for mIoU over the previous state-of-the-art (SOTA) method on the VOC and COCO datasets, respectively, and establishes a new SOTA FID score for conditional image generation amongst slot-attention-based methods. For the segmentation task, GLASS surpasses SOTA weakly-supervised and language-based segmentation models, which were specifically designed for the task.
Authors: Yaxi Chen, Aleksandra Ivanova, Shaheer U. Saeed, Rikin Hargunani, Jie Huang, Chaozong Liu, Yipeng Hu
Abstract: The recently proposed Segment Anything Model (SAM) is a general tool for image segmentation, but it requires additional adaptation and careful fine-tuning for medical image segmentation, especially for small, irregularly-shaped, and boundary-ambiguous anatomical structures such as the knee cartilage that is of interest in this work. Repaired cartilage, after certain surgical procedures, exhibits imaging patterns unseen to pre-training, posing further challenges for using models like SAM with or without general-purpose fine-tuning. To address this, we propose a novel registration-based prompt engineering framework for medical image segmentation using SAM. This approach utilises established image registration algorithms to align the new image (to-be-segmented) and a small number of reference images, without requiring segmentation labels. The spatial transformations generated by registration align either the new image or pre-defined point-based prompts, before using them as input to SAM. This strategy, requiring as few as five reference images with defined point prompts, effectively prompts SAM for inference on new images, without needing any segmentation labels. Evaluation of MR images from patients who received cartilage stem cell therapy yielded Dice scores of 0.89, 0.87, 0.53, and 0.52 for segmenting femur, tibia, femoral- and tibial cartilages, respectively. This outperforms atlas-based label fusion and is comparable to supervised nnUNet, an upper-bound fair baseline in this application, both of which require full segmentation labels for reference samples. The codes are available at: https://github.com/chrissyinreallife/KneeSegmentWithSAM.git
URLs: https://github.com/chrissyinreallife/KneeSegmentWithSAM.git
Authors: Amna Imran, Meghana Shashishekhara Hulikal, Hamza A. A. Gardi
Abstract: This paper focuses on real-time American Sign Language Detection. YOLO is a convolutional neural network (CNN) based model, which was first released in 2015. In recent years, it gained popularity for its real-time detection capabilities. Our study specifically targets YOLO-v9 model, released in 2024. As the model is newly introduced, not much work has been done on it, especially not in Sign Language Detection. Our paper provides deep insight on how YOLO- v9 works and better than previous model.
Authors: Xiang Zhang, Bingxin Ke, Hayko Riemenschneider, Nando Metzger, Anton Obukhov, Markus Gross, Konrad Schindler, Christopher Schroers
Abstract: By training over large-scale datasets, zero-shot monocular depth estimation (MDE) methods show robust performance in the wild but often suffer from insufficiently precise details. Although recent diffusion-based MDE approaches exhibit appealing detail extraction ability, they still struggle in geometrically challenging scenes due to the difficulty of gaining robust geometric priors from diverse datasets. To leverage the complementary merits of both worlds, we propose BetterDepth to efficiently achieve geometrically correct affine-invariant MDE performance while capturing fine-grained details. Specifically, BetterDepth is a conditional diffusion-based refiner that takes the prediction from pre-trained MDE models as depth conditioning, in which the global depth context is well-captured, and iteratively refines details based on the input image. For the training of such a refiner, we propose global pre-alignment and local patch masking methods to ensure the faithfulness of BetterDepth to depth conditioning while learning to capture fine-grained scene details. By efficient training on small-scale synthetic datasets, BetterDepth achieves state-of-the-art zero-shot MDE performance on diverse public datasets and in-the-wild scenes. Moreover, BetterDepth can improve the performance of other MDE models in a plug-and-play manner without additional re-training.
Authors: Katherine L. Mentzer, Andrea Montanari
Abstract: Empirically-determined scaling laws have been broadly successful in predicting the evolution of large machine learning models with training data and number of parameters. As a consequence, they have been useful for optimizing the allocation of limited resources, most notably compute time. In certain applications, storage space is an important constraint, and data format needs to be chosen carefully as a consequence. Computer vision is a prominent example: images are inherently analog, but are always stored in a digital format using a finite number of bits. Given a dataset of digital images, the number of bits $L$ to store each of them can be further reduced using lossy data compression. This, however, can degrade the quality of the model trained on such images, since each example has lower resolution. In order to capture this trade-off and optimize storage of training data, we propose a `storage scaling law' that describes the joint evolution of test error with sample size and number of bits per image. We prove that this law holds within a stylized model for image compression, and verify it empirically on two computer vision tasks, extracting the relevant parameters. We then show that this law can be used to optimize the lossy compression level. At given storage, models trained on optimally compressed images present a significantly smaller test error with respect to models trained on the original data. Finally, we investigate the potential benefits of randomizing the compression level.
Authors: Wenxi Li, Ruxin Zhang, Haozhe Lin, Yuchen Guo, Chao Ma, Xiaokang Yang
Abstract: The advancement of deep learning in object detection has predominantly focused on megapixel images, leaving a critical gap in the efficient processing of gigapixel images. These super high-resolution images present unique challenges due to their immense size and computational demands. To address this, we introduce 'SaccadeDet', an innovative architecture for gigapixel-level object detection, inspired by the human eye saccadic movement. The cornerstone of SaccadeDet is its ability to strategically select and process image regions, dramatically reducing computational load. This is achieved through a two-stage process: the 'saccade' stage, which identifies regions of probable interest, and the 'gaze' stage, which refines detection in these targeted areas. Our approach, evaluated on the PANDA dataset, not only achieves an 8x speed increase over the state-of-the-art methods but also demonstrates significant potential in gigapixel-level pathology analysis through its application to Whole Slide Imaging.
Authors: Kailai Zhou, Lijing Cai, Yibo Wang, Mengya Zhang, Bihan Wen, Qiu Shen, Xun Cao
Abstract: The integration of miniaturized spectrometers into mobile devices offers new avenues for image quality enhancement and facilitates novel downstream tasks. However, the broader application of spectral sensors in mobile photography is hindered by the inherent complexity of spectral images and the constraints of spectral imaging capabilities. To overcome these challenges, we propose a joint RGB-Spectral decomposition model guided enhancement framework, which consists of two steps: joint decomposition and prior-guided enhancement. Firstly, we leverage the complementarity between RGB and Low-resolution Multi-Spectral Images (Lr-MSI) to predict shading, reflectance, and material semantic priors. Subsequently, these priors are seamlessly integrated into the established HDRNet to promote dynamic range enhancement, color mapping, and grid expert learning, respectively. Additionally, we construct a high-quality Mobile-Spec dataset to support our research, and our experiments validate the effectiveness of Lr-MSI in the tone enhancement task. This work aims to establish a solid foundation for advancing spectral vision in mobile photography. The code is available at \url{https://github.com/CalayZhou/JDM-HDRNet}.
Authors: Ryosuke Wayama, Yuki Sasaki, Satoshi Kagiwada, Nobusuke Iwasaki, Hitoshi Iyatomi
Abstract: The development of practical and robust automated diagnostic systems for identifying plant pests is crucial for efficient agricultural production. In this paper, we first investigate three key research questions (RQs) that have not been addressed thus far in the field of image-based plant pest identification. Based on the knowledge gained, we then develop an accurate, robust, and fast plant pest identification framework using 334K images comprising 78 combinations of four plant portions (the leaf front, leaf back, fruit, and flower of cucumber, tomato, strawberry, and eggplant) and 20 pest species captured at 27 farms. The results reveal the following. (1) For an appropriate evaluation of the model, the test data should not include images of the field from which the training images were collected, or other considerations to increase the diversity of the test set should be taken into account. (2) Pre-extraction of ROIs, such as leaves and fruits, helps to improve identification accuracy. (3) Integration of closely related species using the same control methods and cross-crop training methods for the same pests, are effective. Our two-stage plant pest identification framework, enabling ROI detection and convolutional neural network (CNN)-based identification, achieved a highly practical performance of 91.0% and 88.5% in mean accuracy and macro F1 score, respectively, for 12,223 instances of test data of 21 classes collected from unseen fields, where 25 classes of images from 318,971 samples were used for training; the average identification time was 476 ms/image.
Authors: Junho Park, Kyeongbo Kong, Suk-Ju Kang
Abstract: Recently, there has been a significant amount of research conducted on 3D hand reconstruction to use various forms of human-computer interaction. However, 3D hand reconstruction in the wild is challenging due to extreme lack of in-the-wild 3D hand datasets. Especially, when hands are in complex pose such as interacting hands, the problems like appearance similarity, self-handed occclusion and depth ambiguity make it more difficult. To overcome these issues, we propose AttentionHand, a novel method for text-driven controllable hand image generation. Since AttentionHand can generate various and numerous in-the-wild hand images well-aligned with 3D hand label, we can acquire a new 3D hand dataset, and can relieve the domain gap between indoor and outdoor scenes. Our method needs easy-to-use four modalities (i.e, an RGB image, a hand mesh image from 3D label, a bounding box, and a text prompt). These modalities are embedded into the latent space by the encoding phase. Then, through the text attention stage, hand-related tokens from the given text prompt are attended to highlight hand-related regions of the latent embedding. After the highlighted embedding is fed to the visual attention stage, hand-related regions in the embedding are attended by conditioning global and local hand mesh images with the diffusion-based pipeline. In the decoding phase, the final feature is decoded to new hand images, which are well-aligned with the given hand mesh image and text prompt. As a result, AttentionHand achieved state-of-the-art among text-to-hand image generation models, and the performance of 3D hand mesh reconstruction was improved by additionally training with hand images generated by AttentionHand.
Authors: Haoyu Chen, Wenbo Li, Jinjin Gu, Jingjing Ren, Sixiang Chen, Tian Ye, Renjing Pei, Kaiwen Zhou, Fenglong Song, Lei Zhu
Abstract: Natural images captured by mobile devices often suffer from multiple types of degradation, such as noise, blur, and low light. Traditional image restoration methods require manual selection of specific tasks, algorithms, and execution sequences, which is time-consuming and may yield suboptimal results. All-in-one models, though capable of handling multiple tasks, typically support only a limited range and often produce overly smooth, low-fidelity outcomes due to their broad data distribution fitting. To address these challenges, we first define a new pipeline for restoring images with multiple degradations, and then introduce RestoreAgent, an intelligent image restoration system leveraging multimodal large language models. RestoreAgent autonomously assesses the type and extent of degradation in input images and performs restoration through (1) determining the appropriate restoration tasks, (2) optimizing the task sequence, (3) selecting the most suitable models, and (4) executing the restoration. Experimental results demonstrate the superior performance of RestoreAgent in handling complex degradation, surpassing human experts. Furthermore, the system modular design facilitates the fast integration of new tasks and models, enhancing its flexibility and scalability for various applications.
Authors: Guanfeng Tang, Zhiyuan Wu, Rui Fan
Abstract: Semantic segmentation and stereo matching, respectively analogous to the ventral and dorsal streams in our human brain, are two key components of autonomous driving perception systems. Addressing these two tasks with separate networks is no longer the mainstream direction in developing computer vision algorithms, particularly with the recent advances in large vision models and embodied artificial intelligence. The trend is shifting towards combining them within a joint learning framework, especially emphasizing feature sharing between the two tasks. The major contributions of this study lie in comprehensively tightening the coupling between semantic segmentation and stereo matching. Specifically, this study introduces three novelties: (1) a tightly coupled, gated feature fusion strategy, (2) a hierarchical deep supervision strategy, and (3) a coupling tightening loss function. The combined use of these technical contributions results in TiCoSS, a state-of-the-art joint learning framework that simultaneously tackles semantic segmentation and stereo matching. Through extensive experiments on the KITTI and vKITTI2 datasets, along with qualitative and quantitative analyses, we validate the effectiveness of our developed strategies and loss function, and demonstrate its superior performance compared to prior arts, with a notable increase in mIoU by over 9%. Our source code will be publicly available at mias.group/TiCoSS upon publication.
Authors: Jintong Hu, Bin Xia, Bin Chen, Wenming Yang, Lei Zhang
Abstract: Implicit neural representations (INRs) have significantly advanced the field of arbitrary-scale super-resolution (ASSR) of images. Most existing INR-based ASSR networks first extract features from the given low-resolution image using an encoder, and then render the super-resolved result via a multi-layer perceptron decoder. Although these approaches have shown promising results, their performance is constrained by the limited representation ability of discrete latent codes in the encoded features. In this paper, we propose a novel ASSR method named GaussianSR that overcomes this limitation through 2D Gaussian Splatting (2DGS). Unlike traditional methods that treat pixels as discrete points, GaussianSR represents each pixel as a continuous Gaussian field. The encoded features are simultaneously refined and upsampled by rendering the mutually stacked Gaussian fields. As a result, long-range dependencies are established to enhance representation ability. In addition, a classifier is developed to dynamically assign Gaussian kernels to all pixels to further improve flexibility. All components of GaussianSR (i.e., encoder, classifier, Gaussian kernels, and decoder) are jointly learned end-to-end. Experiments demonstrate that GaussianSR achieves superior ASSR performance with fewer parameters than existing methods while enjoying interpretable and content-aware feature aggregations.
Authors: A. Emin Orhan
Abstract: We introduce Human-like Video Models (HVM-1), large-scale video models pretrained with nearly 5000 hours of curated human-like video data (mostly egocentric, temporally extended, continuous video recordings), using the spatiotemporal masked autoencoder (ST-MAE) algorithm. We release two 633M parameter models trained at spatial resolutions of 224x224 and 448x448 pixels. We evaluate the performance of these models in downstream few-shot video and image recognition tasks and compare them against a model pretrained with 1330 hours of short action-oriented video clips from YouTube (Kinetics-700). HVM-1 models perform competitively against the Kinetics-700 pretrained model in downstream evaluations despite substantial qualitative differences between the spatiotemporal characteristics of the corresponding pretraining datasets. HVM-1 models also learn more accurate and more robust object representations compared to models pretrained with the image-based MAE algorithm on the same data, demonstrating the potential benefits of learning to predict temporal regularities in natural videos for learning better object representations.
Authors: Zijian Zhu, Ali Zia, Xuesong Li, Bingbing Dan, Yuebo Ma, Enhai Liu, Rujin Zhao
Abstract: Stripe-like space target detection (SSTD) plays a key role in enhancing space situational awareness and assessing spacecraft behaviour. This domain faces three challenges: the lack of publicly available datasets, interference from stray light and stars, and the variability of stripe-like targets, which complicates pixel-level annotation. In response, we introduces `AstroStripeSet', a pioneering dataset designed for SSTD, aiming to bridge the gap in academic resources and advance research in SSTD. Furthermore, we propose a novel pseudo-label evolution teacher-student framework with single-point supervision. This framework starts with generating initial pseudo-labels using the zero-shot capabilities of the Segment Anything Model (SAM) in a single-point setting, and refines these labels iteratively. In our framework, the fine-tuned StripeSAM serves as the teacher and the newly developed StripeNet as the student, consistently improving segmentation performance by improving the quality of pseudo-labels. We also introduce `GeoDice', a new loss function customized for the linear characteristics of stripe-like targets. Extensive experiments show that the performance of our approach matches fully supervised methods on all evaluation metrics, establishing a new state-of-the-art (SOTA) benchmark. Our dataset and code will be made publicly available.
Authors: Florent Brondolo, Samuel Beaussant
Abstract: This study investigates the interpretability, classification, and segmentation of CT-scan images of rock samples, with a particular focus on the application of DINOv2 within Geosciences. We compared various segmentation techniques to evaluate their efficacy, efficiency, and adaptability in geological image analysis. The methods assessed include the Otsu thresholding method, clustering techniques (K-means and fuzzy C-means), a supervised machine learning approach (Random Forest), and deep learning methods (UNet and DINOv2). We tested these methods using ten binary sandstone datasets and three multi-class calcite datasets. To begin, we provide a thorough interpretability analysis of DINOv2's features in the geoscientific context, discussing its suitability and inherent ability to process CT-scanned rock data. In terms of classification, the out-of-the-box DINOv2 demonstrates an impressive capability to perfectly classify rock images, even when the CT scans are out of its original training set. Regarding segmentation, thresholding and unsupervised methods, while fast, perform poorly despite image preprocessing, whereas supervised methods show better results. We underscore the computational demands of deep learning but highlight its minimal intervention, superior generalization, and performance without additional image preprocessing. Additionally, we observe a lack of correlation between a network's depth or the number of parameters and its performance. Our results show that a LoRA fine-tuned DINOv2 excels in out-of-distribution segmentation and significantly outperforms other methods in multi-class segmentation. By systematically comparing these methods, we identify the most efficient strategy for meticulous and laborious segmentation tasks. DINOv2 proves advantageous, achieving segmentations that could be described as "better than ground-truth" against relatively small training sets.
Authors: Vladimir Somers, Christophe De Vleeschouwer, Alexandre Alahi
Abstract: Occluded Person Re-Identification (ReID) is a metric learning task that involves matching occluded individuals based on their appearance. While many studies have tackled occlusions caused by objects, multi-person occlusions remain less explored. In this work, we identify and address a critical challenge overlooked by previous occluded ReID methods: the Multi-Person Ambiguity (MPA) arising when multiple individuals are visible in the same bounding box, making it impossible to determine the intended ReID target among the candidates. Inspired by recent work on prompting in vision, we introduce Keypoint Promptable ReID (KPR), a novel formulation of the ReID problem that explicitly complements the input bounding box with a set of semantic keypoints indicating the intended target. Since promptable re-identification is an unexplored paradigm, existing ReID datasets lack the pixel-level annotations necessary for prompting. To bridge this gap and foster further research on this topic, we introduce Occluded-PoseTrack ReID, a novel ReID dataset with keypoints labels, that features strong inter-person occlusions. Furthermore, we release custom keypoint labels for four popular ReID benchmarks. Experiments on person retrieval, but also on pose tracking, demonstrate that our method systematically surpasses previous state-of-the-art approaches on various occluded scenarios. Our code, dataset and annotations are available at https://github.com/VlSomers/keypoint_promptable_reidentification.
URLs: https://github.com/VlSomers/keypoint_promptable_reidentification.
Authors: Zuyan Liu, Benlin Liu, Jiahui Wang, Yuhao Dong, Guangyi Chen, Yongming Rao, Ranjay Krishna, Jiwen Lu
Abstract: In the field of instruction-following large vision-language models (LVLMs), the efficient deployment of these models faces challenges, notably due to the high memory demands of their key-value (KV) caches. Conventional cache management strategies for LLMs focus on cache eviction, which often fails to address the specific needs of multimodal instruction-following models. Recognizing this gap, in this paper, we introduce Elastic Cache, a novel approach that benefits from applying distinct acceleration methods for instruction encoding and output generation stages. We investigate the metrics of importance in different stages and propose an importance-driven cache merging strategy to prune redundancy caches. Instead of discarding less important caches, our strategy identifies important key/value vectors as anchor points. Surrounding less important caches are then merged with these anchors, enhancing the preservation of contextual information in the KV caches while yielding an arbitrary acceleration ratio. For instruction encoding, we utilize the frequency to evaluate the importance of caches. Regarding output generation, we prioritize tokens based on their distance with an offset, by which both the initial and most recent tokens are retained. Results on a range of LVLMs demonstrate that Elastic Cache not only boosts efficiency but also notably outperforms existing pruning methods in language generation across various tasks. Code is available at https://github.com/liuzuyan/ElasticCache
Authors: Roberto Di Via, Francesca Odone, Vito Paolo Pastore
Abstract: In the last few years, deep neural networks have been extensively applied in the medical domain for different tasks, ranging from image classification and segmentation to landmark detection. However, the application of these technologies in the medical domain is often hindered by data scarcity, both in terms of available annotations and images. This study introduces a new self-supervised pre-training protocol based on diffusion models for landmark detection in x-ray images. Our results show that the proposed self-supervised framework can provide accurate landmark detection with a minimal number of available annotated training images (up to 50), outperforming ImageNet supervised pre-training and state-of-the-art self-supervised pre-trainings for three popular x-ray benchmark datasets. To our knowledge, this is the first exploration of diffusion models for self-supervised learning in landmark detection, which may offer a valuable pre-training approach in few-shot regimes, for mitigating data scarcity.
Authors: Daniele Rege Cambrin, Isaac Corley, Paolo Garza, Peyman Najafirad
Abstract: Earthquakes are commonly estimated using physical seismic stations, however, due to the installation requirements and costs of these stations, global coverage quickly becomes impractical. An efficient and lower-cost alternative is to develop machine learning models to globally monitor earth observation data to pinpoint regions impacted by these natural disasters. However, due to the small amount of historically recorded earthquakes, this becomes a low-data regime problem requiring algorithmic improvements to achieve peak performance when learning to regress earthquake magnitude. In this paper, we propose to pose the estimation of earthquake magnitudes as a metric-learning problem, training models to not only estimate earthquake magnitude from Sentinel-1 satellite imagery but to additionally rank pairwise samples. Our experiments show at max a 30%+ improvement in MAE over prior regression-only based methods, particularly transformer-based architectures.
Authors: Vlad Sobal, Mark Ibrahim, Randall Balestriero, Vivien Cabannes, Diane Bouchacourt, Pietro Astolfi, Kyunghyun Cho, Yann LeCun
Abstract: Learning good representations involves capturing the diverse ways in which data samples relate. Contrastive loss - an objective matching related samples - underlies methods from self-supervised to multimodal learning. Contrastive losses, however, can be viewed more broadly as modifying a similarity graph to indicate how samples should relate in the embedding space. This view reveals a shortcoming in contrastive learning: the similarity graph is binary, as only one sample is the related positive sample. Crucially, similarities \textit{across} samples are ignored. Based on this observation, we revise the standard contrastive loss to explicitly encode how a sample relates to others. We experiment with this new objective, called $\mathbb{X}$-Sample Contrastive, to train vision models based on similarities in class or text caption descriptions. Our study spans three scales: ImageNet-1k with 1 million, CC3M with 3 million, and CC12M with 12 million samples. The representations learned via our objective outperform both contrastive self-supervised and vision-language models trained on the same data across a range of tasks. When training on CC12M, we outperform CLIP by $0.6\%$ on both ImageNet and ImageNet Real. Our objective appears to work particularly well in lower-data regimes, with gains over CLIP of $16.8\%$ on ImageNet and $18.1\%$ on ImageNet Real when training with CC3M. Finally, our objective seems to encourage the model to learn representations that separate objects from their attributes and backgrounds, with gains of $3.3$-$5.6$\% over CLIP on ImageNet9. We hope the proposed solution takes a small step towards developing richer learning objectives for understanding sample relations in foundation models.
Authors: Jiahao Guo, Ziyang Xu, Lianjun Wu, Fei Gao, Wenyu Liu, Xinggang Wang
Abstract: Small Video Object Detection (SVOD) is a crucial subfield in modern computer vision, essential for early object discovery and detection. However, existing SVOD datasets are scarce and suffer from issues such as insufficiently small objects, limited object categories, and lack of scene diversity, leading to unitary application scenarios for corresponding methods. To address this gap, we develop the XS-VID dataset, which comprises aerial data from various periods and scenes, and annotates eight major object categories. To further evaluate existing methods for detecting extremely small objects, XS-VID extensively collects three types of objects with smaller pixel areas: extremely small (\textit{es}, $0\sim12^2$), relatively small (\textit{rs}, $12^2\sim20^2$), and generally small (\textit{gs}, $20^2\sim32^2$). XS-VID offers unprecedented breadth and depth in covering and quantifying minuscule objects, significantly enriching the scene and object diversity in the dataset. Extensive validations on XS-VID and the publicly available VisDrone2019VID dataset show that existing methods struggle with small object detection and significantly underperform compared to general object detectors. Leveraging the strengths of previous methods and addressing their weaknesses, we propose YOLOFT, which enhances local feature associations and integrates temporal motion features, significantly improving the accuracy and stability of SVOD. Our datasets and benchmarks are available at \url{https://gjhhust.github.io/XS-VID/}.
Authors: Julia Hindel, Daniele Cattaneo, Abhinav Valada
Abstract: Semantic segmentation models are typically trained on a fixed set of classes, limiting their applicability in open-world scenarios. Class-incremental semantic segmentation aims to update models with emerging new classes while preventing catastrophic forgetting of previously learned ones. However, existing methods impose strict rigidity on old classes, reducing their effectiveness in learning new incremental classes. In this work, we propose Taxonomy-Oriented Poincar\'e-regularized Incremental-Class Segmentation (TOPICS) that learns feature embeddings in hyperbolic space following explicit taxonomy-tree structures. This supervision provides plasticity for old classes, updating ancestors based on new classes while integrating new classes at fitting positions. Additionally, we maintain implicit class relational constraints on the geometric basis of the Poincar\'e ball. This ensures that the latent space can continuously adapt to new constraints while maintaining a robust structure to combat catastrophic forgetting. We also establish eight realistic incremental learning protocols for autonomous driving scenarios, where novel classes can originate from known classes or the background. Extensive evaluations of TOPICS on the Cityscapes and Mapillary Vistas 2.0 benchmarks demonstrate that it achieves state-of-the-art performance. We make the code and trained models publicly available at http://topics.cs.uni-freiburg.de.
Authors: Cheng Qian, Julen Urain, Kevin Zakka, Jan Peters
Abstract: In this work, we introduce PianoMime, a framework for training a piano-playing agent using internet demonstrations. The internet is a promising source of large-scale demonstrations for training our robot agents. In particular, for the case of piano-playing, Youtube is full of videos of professional pianists playing a wide myriad of songs. In our work, we leverage these demonstrations to learn a generalist piano-playing agent capable of playing any arbitrary song. Our framework is divided into three parts: a data preparation phase to extract the informative features from the Youtube videos, a policy learning phase to train song-specific expert policies from the demonstrations and a policy distillation phase to distil the policies into a single generalist agent. We explore different policy designs to represent the agent and evaluate the influence of the amount of training data on the generalization capability of the agent to novel songs not available in the dataset. We show that we are able to learn a policy with up to 56\% F1 score on unseen songs.
Authors: Anders Christensen, Nooshin Mojab, Khushman Patel, Karan Ahuja, Zeynep Akata, Ole Winther, Mar Gonzalez-Franco, Andrea Colaco
Abstract: Spherical or omni-directional images offer an immersive visual format appealing to a wide range of computer vision applications. However, geometric properties of spherical images pose a major challenge for models and metrics designed for ordinary 2D images. Here, we show that direct application of Fr\'echet Inception Distance (FID) is insufficient for quantifying geometric fidelity in spherical images. We introduce two quantitative metrics accounting for geometric constraints, namely Omnidirectional FID (OmniFID) and Discontinuity Score (DS). OmniFID is an extension of FID tailored to additionally capture field-of-view requirements of the spherical format by leveraging cubemap projections. DS is a kernel-based seam alignment score of continuity across borders of 2D representations of spherical images. In experiments, OmniFID and DS quantify geometry fidelity issues that are undetected by FID.
Authors: Zhe Liu, Jinghua Hou, Xinyu Wang, Xiaoqing Ye, Jingdong Wang, Hengshuang Zhao, Xiang Bai
Abstract: The benefit of transformers in large-scale 3D point cloud perception tasks, such as 3D object detection, is limited by their quadratic computation cost when modeling long-range relationships. In contrast, linear RNNs have low computational complexity and are suitable for long-range modeling. Toward this goal, we propose a simple and effective window-based framework built on LInear grOup RNN (i.e., perform linear RNN for grouped features) for accurate 3D object detection, called LION. The key property is to allow sufficient feature interaction in a much larger group than transformer-based methods. However, effectively applying linear group RNN to 3D object detection in highly sparse point clouds is not trivial due to its limitation in handling spatial modeling. To tackle this problem, we simply introduce a 3D spatial feature descriptor and integrate it into the linear group RNN operators to enhance their spatial features rather than blindly increasing the number of scanning orders for voxel features. To further address the challenge in highly sparse point clouds, we propose a 3D voxel generation strategy to densify foreground features thanks to linear group RNN as a natural property of auto-regressive models. Extensive experiments verify the effectiveness of the proposed components and the generalization of our LION on different linear group RNN operators including Mamba, RWKV, and RetNet. Furthermore, it is worth mentioning that our LION-Mamba achieves state-of-the-art on Waymo, nuScenes, Argoverse V2, and ONCE dataset. Last but not least, our method supports kinds of advanced linear RNN operators (e.g., RetNet, RWKV, Mamba, xLSTM and TTT) on small but popular KITTI dataset for a quick experience with our linear RNN-based framework.
Authors: Yu-Yun Tseng, Tanusree Sharma, Lotus Zhang, Abigale Stangl, Leah Findlater, Yang Wang, Danna Gurari Yu-Yun Tseng, Tanusree Sharma, Lotus Zhang, Abigale Stangl, Leah Findlater, Yang Wang, Danna Gurari
Abstract: Individuals who are blind or have low vision (BLV) are at a heightened risk of sharing private information if they share photographs they have taken. To facilitate developing technologies that can help preserve privacy, we introduce BIV-Priv-Seg, the first localization dataset originating from people with visual impairments that shows private content. It contains 1,028 images with segmentation annotations for 16 private object categories. We first characterize BIV-Priv-Seg and then evaluate modern models' performance for locating private content in the dataset. We find modern models struggle most with locating private objects that are not salient, small, and lack text as well as recognizing when private content is absent from an image. We facilitate future extensions by sharing our new dataset with the evaluation server at https://vizwiz.org/tasks-and-datasets/object-localization.
URLs: https://vizwiz.org/tasks-and-datasets/object-localization.
Authors: Shuting He, Henghui Ding
Abstract: 3D referring segmentation is an emerging and challenging vision-language task that aims to segment the object described by a natural language expression in a point cloud scene. The key challenge behind this task is vision-language feature fusion and alignment. In this work, we propose RefMask3D to explore the comprehensive multi-modal feature interaction and understanding. First, we propose a Geometry-Enhanced Group-Word Attention to integrate language with geometrically coherent sub-clouds through cross-modal group-word attention, which effectively addresses the challenges posed by the sparse and irregular nature of point clouds. Then, we introduce a Linguistic Primitives Construction to produce semantic primitives representing distinct semantic attributes, which greatly enhance the vision-language understanding at the decoding stage. Furthermore, we introduce an Object Cluster Module that analyzes the interrelationships among linguistic primitives to consolidate their insights and pinpoint common characteristics, helping to capture holistic information and enhance the precision of target identification. The proposed RefMask3D achieves new state-of-the-art performance on 3D referring segmentation, 3D visual grounding, and also 2D referring image segmentation. Especially, RefMask3D outperforms previous state-of-the-art method by a large margin of 3.16% mIoU} on the challenging ScanRefer dataset. Code is available at https://github.com/heshuting555/RefMask3D.
Authors: Orest Kupyn, Eugene Khvedchenia, Christian Rupprecht
Abstract: Human head detection, keypoint estimation, and 3D head model fitting are important tasks with many applications. However, traditional real-world datasets often suffer from bias, privacy, and ethical concerns, and they have been recorded in laboratory environments, which makes it difficult for trained models to generalize. Here, we introduce VGGHeads -- a large scale synthetic dataset generated with diffusion models for human head detection and 3D mesh estimation. Our dataset comprises over 1 million high-resolution images, each annotated with detailed 3D head meshes, facial landmarks, and bounding boxes. Using this dataset we introduce a new model architecture capable of simultaneous heads detection and head meshes reconstruction from a single image in a single step. Through extensive experimental evaluations, we demonstrate that models trained on our synthetic data achieve strong performance on real images. Furthermore, the versatility of our dataset makes it applicable across a broad spectrum of tasks, offering a general and comprehensive representation of human heads. Additionally, we provide detailed information about the synthetic data generation pipeline, enabling it to be re-used for other tasks and domains.
Authors: Jingyi Lu, Xinghui Li, Kai Han
Abstract: Point-drag-based image editing methods, like DragDiffusion, have attracted significant attention. However, point-drag-based approaches suffer from computational overhead and misinterpretation of user intentions due to the sparsity of point-based editing instructions. In this paper, we propose a region-based copy-and-paste dragging method, RegionDrag, to overcome these limitations. RegionDrag allows users to express their editing instructions in the form of handle and target regions, enabling more precise control and alleviating ambiguity. In addition, region-based operations complete editing in one iteration and are much faster than point-drag-based methods. We also incorporate the attention-swapping technique for enhanced stability during editing. To validate our approach, we extend existing point-drag-based datasets with region-based dragging instructions. Experimental results demonstrate that RegionDrag outperforms existing point-drag-based approaches in terms of speed, accuracy, and alignment with user intentions. Remarkably, RegionDrag completes the edit on an image with a resolution of 512x512 in less than 2 seconds, which is more than 100x faster than DragDiffusion, while achieving better performance. Project page: https://visual-ai.github.io/regiondrag.
Authors: Pulkit Kumar, Namitha Padmanabhan, Luke Luo, Sai Saketh Rambhatla, Abhinav Shrivastava
Abstract: We propose a simple yet effective approach for few-shot action recognition, emphasizing the disentanglement of motion and appearance representations. By harnessing recent progress in tracking, specifically point trajectories and self-supervised representation learning, we build trajectory-aligned tokens (TATs) that capture motion and appearance information. This approach significantly reduces the data requirements while retaining essential information. To process these representations, we use a Masked Space-time Transformer that effectively learns to aggregate information to facilitate few-shot action recognition. We demonstrate state-of-the-art results on few-shot action recognition across multiple datasets. Our project page is available at https://www.cs.umd.edu/~pulkit/tats
URLs: https://www.cs.umd.edu/
Authors: Cristian-Alexandru Botocan, Raphael Meier, Ljiljana Dolamic
Abstract: Assessing the robustness of multimodal models against adversarial examples is an important aspect for the safety of its users. We craft L0-norm perturbation attacks on the preprocessed input images. We launch them in a black-box setup against four multimodal models and two unimodal DNNs, considering both targeted and untargeted misclassification. Our attacks target less than 0.04% of perturbed image area and integrate different spatial positioning of perturbed pixels: sparse positioning and pixels arranged in different contiguous shapes (row, column, diagonal, and patch). To the best of our knowledge, we are the first to assess the robustness of three state-of-the-art multimodal models (ALIGN, AltCLIP, GroupViT) against different sparse and contiguous pixel distribution perturbations. The obtained results indicate that unimodal DNNs are more robust than multimodal models. Furthermore, models using CNN-based Image Encoder are more vulnerable than models with ViT - for untargeted attacks, we obtain a 99% success rate by perturbing less than 0.02% of the image area.
Authors: Linhao Qu, Dan Huang, Shaoting Zhang, Xiaosong Wang
Abstract: Survival analysis stands as a pivotal process in cancer treatment research, crucial for predicting patient survival rates accurately. Recent advancements in data collection techniques have paved the way for enhancing survival predictions by integrating information from multiple modalities. However, real-world scenarios often present challenges with incomplete data, particularly when dealing with censored survival labels. Prior works have addressed missing modalities but have overlooked incomplete labels, which can introduce bias and limit model efficacy. To bridge this gap, we introduce a novel framework that simultaneously handles incomplete data across modalities and censored survival labels. Our approach employs advanced foundation models to encode individual modalities and align them into a universal representation space for seamless fusion. By generating pseudo labels and incorporating uncertainty, we significantly enhance predictive accuracy. The proposed method demonstrates outstanding prediction accuracy in two survival analysis tasks on both employed datasets. This innovative approach overcomes limitations associated with disparate modalities and improves the feasibility of comprehensive survival analysis using multiple large foundation models.
Authors: Kaitao Chen, Mianxin Liu, Fang Yan, Lei Ma, Xiaoming Shi, Lilong Wang, Xiaosong Wang, Lifeng Zhu, Zhe Wang, Mu Zhou, Shaoting Zhang
Abstract: The advent of vision-language models fosters the interactive conversations between AI-enabled models and humans. Yet applying these models into clinics must deal with daunting challenges around large-scale training data, financial, and computational resources. Here we propose a cost-effective instruction learning framework for conversational pathology named as CLOVER. CLOVER only trains a lightweight module and uses instruction tuning while freezing the parameters of the large language model. Instead of using costly GPT-4, we propose well-designed prompts on GPT-3.5 for building generation-based instructions, emphasizing the utility of pathological knowledge derived from the Internet source. To augment the use of instructions, we construct a high-quality set of template-based instructions in the context of digital pathology. From two benchmark datasets, our findings reveal the strength of hybrid-form instructions in the visual question-answer in pathology. Extensive results show the cost-effectiveness of CLOVER in answering both open-ended and closed-ended questions, where CLOVER outperforms strong baselines that possess 37 times more training parameters and use instruction data generated from GPT-4. Through the instruction tuning, CLOVER exhibits robustness of few-shot learning in the external clinical dataset. These findings demonstrate that cost-effective modeling of CLOVER could accelerate the adoption of rapid conversational applications in the landscape of digital pathology.
Authors: Tajamul Ashraf, Tisha Madame
Abstract: In clinical applications, X-ray technology is vital for noninvasive examinations like mammography, providing essential anatomical information. However, the radiation risk associated with X-ray procedures raises concerns. X-ray reconstruction is crucial in medical imaging for detailed visual representations of internal structures, aiding diagnosis and treatment without invasive procedures. Recent advancements in deep learning (DL) have shown promise in X-ray reconstruction, but conventional DL methods often require centralized aggregation of large datasets, leading to domain shifts and privacy issues. To address these challenges, we introduce the Hierarchical Framework-based Federated Learning method (HF-Fed) for customized X-ray imaging. HF-Fed tackles X-ray imaging optimization by decomposing the problem into local data adaptation and holistic X-ray imaging. It employs a hospital-specific hierarchical framework and a shared common imaging network called Network of Networks (NoN) to acquire stable features from diverse data distributions. The hierarchical hypernetwork extracts domain-specific hyperparameters, conditioning the NoN for customized X-ray reconstruction. Experimental results demonstrate HF-Fed's competitive performance, offering a promising solution for enhancing X-ray imaging without data sharing. This study significantly contributes to the literature on federated learning in healthcare, providing valuable insights for policymakers and healthcare providers. The source code and pre-trained HF-Fed model are available at \url{https://tisharepo.github.io/Webpage/}.
Authors: Tomer Barak, Yonatan Loewenstein
Abstract: Humans are capable of solving complex abstract reasoning tests. Whether this ability reflects a learning-independent inference mechanism applicable to any novel unlearned problem or whether it is a manifestation of extensive training throughout life is an open question. Addressing this question in humans is challenging because it is impossible to control their prior training. However, assuming a similarity between the cognitive processing of Artificial Neural Networks (ANNs) and humans, the extent to which training is required for ANNs' abstract reasoning is informative about this question in humans. Previous studies demonstrated that ANNs can solve abstract reasoning tests. However, this success required extensive training. In this study, we examined the learning-independent abstract reasoning of ANNs. Specifically, we evaluated their performance without any pretraining, with the ANNs' weights being randomly-initialized, and only change in the process of problem solving. We found that naive ANN models can solve non-trivial visual reasoning tests, similar to those used to evaluate human learning-independent reasoning. We further studied the mechanisms that support this ability. Our results suggest the possibility of learning-independent abstract reasoning that does not require extensive training.
Authors: Jian-Yu Jiang-Lin, Kang-Yang Huang, Ling Lo, Yi-Ning Huang, Terence Lin, Jhih-Ciang Wu, Hong-Han Shuai, Wen-Huang Cheng
Abstract: Diffusion models revolutionize image generation by leveraging natural language to guide the creation of multimedia content. Despite significant advancements in such generative models, challenges persist in depicting detailed human-object interactions, especially regarding pose and object placement accuracy. We introduce a training-free method named Reasoning and Correcting Diffusion (ReCorD) to address these challenges. Our model couples Latent Diffusion Models with Visual Language Models to refine the generation process, ensuring precise depictions of HOIs. We propose an interaction-aware reasoning module to improve the interpretation of the interaction, along with an interaction correcting module to refine the output image for more precise HOI generation delicately. Through a meticulous process of pose selection and object positioning, ReCorD achieves superior fidelity in generated images while efficiently reducing computational requirements. We conduct comprehensive experiments on three benchmarks to demonstrate the significant progress in solving text-to-image generation tasks, showcasing ReCorD's ability to render complex interactions accurately by outperforming existing methods in HOI classification score, as well as FID and Verb CLIP-Score. Project website is available at https://alberthkyhky.github.io/ReCorD/ .
Authors: Debanjali Bhattacharya, Ninad Aithal, Manish Jayswal, Neelam Sinha
Abstract: Recent advances in molecular and genetic research have identified a diverse range of brain tumor sub-types, shedding light on differences in their molecular mechanisms, heterogeneity, and origins. The present study performs whole-brain connectome analysis using diffusionweighted images. To achieve this, both graph theory and persistent homology - a prominent approach in topological data analysis are employed in order to quantify changes in the structural connectivity of the wholebrain connectome in subjects with brain tumors. Probabilistic tractography is used to map the number of streamlines connecting 84 distinct brain regions, as delineated by the Desikan-Killiany atlas from FreeSurfer. These streamline mappings form the connectome matrix, on which persistent homology based analysis and graph theoretical analysis are executed to evaluate the discriminatory power between tumor sub-types that include meningioma and glioma. A detailed statistical analysis is conducted on persistent homology-derived topological features and graphical features to identify the brain regions where differences between study groups are statistically significant (p < 0.05). For classification purpose, graph-based local features are utilized, achieving a highest accuracy of 88%. In classifying tumor sub-types, an accuracy of 80% is attained. The findings obtained from this study underscore the potential of persistent homology and graph theoretical analysis of the whole-brain connectome in detecting alterations in structural connectivity patterns specific to different types of brain tumors.
Authors: Nghia Nguyen, Minh Nhat Vu, Baoru Huang, An Vuong, Ngan Le, Thieu Vo, Anh Nguyen
Abstract: Language-driven grasp detection is a fundamental yet challenging task in robotics with various industrial applications. In this work, we present a new approach for language-driven grasp detection that leverages the concept of lightweight diffusion models to achieve fast inference time. By integrating diffusion processes with grasping prompts in natural language, our method can effectively encode visual and textual information, enabling more accurate and versatile grasp positioning that aligns well with the text query. To overcome the long inference time problem in diffusion models, we leverage the image and text features as the condition in the consistency model to reduce the number of denoising timesteps during inference. The intensive experimental results show that our method outperforms other recent grasp detection methods and lightweight diffusion models by a clear margin. We further validate our method in real-world robotic experiments to demonstrate its fast inference time capability.
Authors: Pirzada Suhail, Amit Sethi
Abstract: Neural networks have emerged as powerful tools across various applications, yet their decision-making process often remains opaque, leading to them being perceived as "black boxes." This opacity raises concerns about their interpretability and reliability, especially in safety-critical scenarios. Network inversion techniques offer a solution by allowing us to peek inside these black boxes, revealing the features and patterns learned by the networks behind their decision-making processes and thereby provide valuable insights into how neural networks arrive at their conclusions, making them more interpretable and trustworthy. This paper presents a simple yet effective approach to network inversion using a carefully conditioned generator that learns the data distribution in the input space of the trained neural network, enabling the reconstruction of inputs that would most likely lead to the desired outputs. To capture the diversity in the input space for a given output, instead of simply revealing the conditioning labels to the generator, we hideously encode the conditioning label information into vectors, further exemplified by heavy dropout in the generation process and minimisation of cosine similarity between the features corresponding to the generated images. The paper concludes with immediate applications of Network Inversion including in interpretability, explainability and generation of adversarial samples.
Authors: Jan Nikolas Morshuis, Matthias Hein, Christian F. Baumgartner
Abstract: Inverse problems, such as accelerated MRI reconstruction, are ill-posed and an infinite amount of possible and plausible solutions exist. This may not only lead to uncertainty in the reconstructed image but also in downstream tasks such as semantic segmentation. This uncertainty, however, is mostly not analyzed in the literature, even though probabilistic reconstruction models are commonly used. These models can be prone to ignore plausible but unlikely solutions like rare pathologies. Building on MRI reconstruction approaches based on diffusion models, we add guidance to the diffusion process during inference, generating two meaningfully diverse reconstructions corresponding to an upper and lower bound segmentation. The reconstruction uncertainty can then be quantified by the difference between these bounds, which we coin the 'uncertainty boundary'. We analyzed the behavior of the upper and lower bound segmentations for a wide range of acceleration factors and found the uncertainty boundary to be both more reliable and more accurate compared to repeated sampling. Code is available at https://github.com/NikolasMorshuis/SGR
Authors: Tianle Zeng, Dengke He, Feifan Yan, Meixi He
Abstract: In a multi-sensor fusion system composed of cameras and LiDAR, precise extrinsic calibration contributes to the system's long-term stability and accurate perception of the environment. However, methods based on extracting and registering corresponding points still face challenges in terms of automation and precision. This paper proposes a novel fully automatic extrinsic calibration method for LiDAR-camera systems that circumvents the need for corresponding point registration. In our approach, a novel algorithm to extract required LiDAR correspondence point is proposed. This method can effectively filter out irrelevant points by computing the orientation of plane point clouds and extracting points by applying distance- and density-based thresholds. We avoid the need for corresponding point registration by introducing extrinsic parameters between the LiDAR and camera into the projection of extracted points and constructing co-planar constraints. These parameters are then optimized to solve for the extrinsic. We validated our method across multiple sets of LiDAR-camera systems. In synthetic experiments, our method demonstrates superior performance compared to current calibration techniques. Real-world data experiments further confirm the precision and robustness of the proposed algorithm, with average rotation and translation calibration errors between LiDAR and camera of less than 0.05 degree and 0.015m, respectively. This method enables automatic and accurate extrinsic calibration in a single one step, emphasizing the potential of calibration algorithms beyond using corresponding point registration to enhance the automation and precision of LiDAR-camera system calibration.
Authors: Ziwei Cui, Jingfeng Yao, Lunbin Zeng, Juan Yang, Wenyu Liu, Xinggang Wang
Abstract: The segmentation of cell nuclei in tissue images stained with the blood dye hematoxylin and eosin (H$\&$E) is essential for various clinical applications and analyses. Due to the complex characteristics of cellular morphology, a large receptive field is considered crucial for generating high-quality segmentation. However, previous methods face challenges in achieving a balance between the receptive field and computational burden. To address this issue, we propose LKCell, a high-accuracy and efficient cell segmentation method. Its core insight lies in unleashing the potential of large convolution kernels to achieve computationally efficient large receptive fields. Specifically, (1) We transfer pre-trained large convolution kernel models to the medical domain for the first time, demonstrating their effectiveness in cell segmentation. (2) We analyze the redundancy of previous methods and design a new segmentation decoder based on large convolution kernels. It achieves higher performance while significantly reducing the number of parameters. We evaluate our method on the most challenging benchmark and achieve state-of-the-art results (0.5080 mPQ) in cell nuclei instance segmentation with only 21.6% FLOPs compared with the previous leading method. Our source code and models are available at https://github.com/hustvl/LKCell.
Authors: Xiao Liu, Peng Gao, Tao Yu, Fei Wang, Ru-Yue Yuan
Abstract: Deep learning, especially convolutional neural networks (CNNs) and Transformer architectures, have become the focus of extensive research in medical image segmentation, achieving impressive results. However, CNNs come with inductive biases that limit their effectiveness in more complex, varied segmentation scenarios. Conversely, while Transformer-based methods excel at capturing global and long-range semantic details, they suffer from high computational demands. In this study, we propose CSWin-UNet, a novel U-shaped segmentation method that incorporates the CSWin self-attention mechanism into the UNet to facilitate horizontal and vertical stripes self-attention. This method significantly enhances both computational efficiency and receptive field interactions. Additionally, our innovative decoder utilizes a content-aware reassembly operator that strategically reassembles features, guided by predicted kernels, for precise image resolution restoration. Our extensive empirical evaluations on diverse datasets, including synapse multi-organ CT, cardiac MRI, and skin lesions, demonstrate that CSWin-UNet maintains low model complexity while delivering high segmentation accuracy.
Authors: Jack Breen, Katie Allen, Kieran Zucker, Nicolas M. Orsi, Nishant Ravikumar
Abstract: Computer vision models are increasingly capable of classifying ovarian epithelial cancer subtypes, but they differ from pathologists by processing small tissue patches at a single resolution. Multi-resolution graph models leverage the spatial relationships of patches at multiple magnifications, learning the context for each patch. In this study, we conduct the most thorough validation of a graph model for ovarian cancer subtyping to date. Seven models were tuned and trained using five-fold cross-validation on a set of 1864 whole slide images (WSIs) from 434 patients treated at Leeds Teaching Hospitals NHS Trust. The cross-validation models were ensembled and evaluated using a balanced hold-out test set of 100 WSIs from 30 patients, and an external validation set of 80 WSIs from 80 patients in the Transcanadian Study. The best-performing model, a graph model using 10x+20x magnification data, gave balanced accuracies of 73%, 88%, and 99% in cross-validation, hold-out testing, and external validation, respectively. However, this only exceeded the performance of attention-based multiple instance learning in external validation, with a 93% balanced accuracy. Graph models benefitted greatly from using the UNI foundation model rather than an ImageNet-pretrained ResNet50 for feature extraction, with this having a much greater effect on performance than changing the subsequent classification approach. The accuracy of the combined foundation model and multi-resolution graph network offers a step towards the clinical applicability of these models, with a new highest-reported performance for this task, though further validations are still required to ensure the robustness and usability of the models.
Authors: Zhengang Li, Alec Lu, Yanyue Xie, Zhenglun Kong, Mengshu Sun, Hao Tang, Zhong Jia Xue, Peiyan Dong, Caiwen Ding, Yanzhi Wang, Xue Lin, Zhenman Fang
Abstract: Vision transformers (ViTs) have demonstrated their superior accuracy for computer vision tasks compared to convolutional neural networks (CNNs). However, ViT models are often computation-intensive for efficient deployment on resource-limited edge devices. This work proposes Quasar-ViT, a hardware-oriented quantization-aware architecture search framework for ViTs, to design efficient ViT models for hardware implementation while preserving the accuracy. First, Quasar-ViT trains a supernet using our row-wise flexible mixed-precision quantization scheme, mixed-precision weight entanglement, and supernet layer scaling techniques. Then, it applies an efficient hardware-oriented search algorithm, integrated with hardware latency and resource modeling, to determine a series of optimal subnets from supernet under different inference latency targets. Finally, we propose a series of model-adaptive designs on the FPGA platform to support the architecture search and mitigate the gap between the theoretical computation reduction and the practical inference speedup. Our searched models achieve 101.5, 159.6, and 251.6 frames-per-second (FPS) inference speed on the AMD/Xilinx ZCU102 FPGA with 80.4%, 78.6%, and 74.9% top-1 accuracy, respectively, for the ImageNet dataset, consistently outperforming prior works.
Authors: Sachin Shah, Naitri Rajyaguru, Chahat Deep Singh, Christopher Metzler, Yiannis Aloimonos
Abstract: Autonomous robots often rely on monocular cameras for odometry estimation and navigation. However, the scale ambiguity problem presents a critical barrier to effective monocular visual odometry. In this paper, we present CodedVO, a novel monocular visual odometry method that overcomes the scale ambiguity problem by employing custom optics to physically encode metric depth information into imagery. By incorporating this information into our odometry pipeline, we achieve state-of-the-art performance in monocular visual odometry with a known scale. We evaluate our method in diverse indoor environments and demonstrate its robustness and adaptability. We achieve a 0.08m average trajectory error in odometry evaluation on the ICL-NUIM indoor odometry dataset.
Authors: Haoran You, Yunyang Xiong, Xiaoliang Dai, Bichen Wu, Peizhao Zhang, Haoqi Fan, Peter Vajda, Yingyan Celine Lin
Abstract: Vision Transformers (ViTs) have shown impressive performance but still require a high computation cost as compared to convolutional neural networks (CNNs), one reason is that ViTs' attention measures global similarities and thus has a quadratic complexity with the number of input tokens. Existing efficient ViTs adopt local attention (e.g., Swin) or linear attention (e.g., Performer), which sacrifice ViTs' capabilities of capturing either global or local context. In this work, we ask an important research question: Can ViTs learn both global and local context while being more efficient during inference? To this end, we propose a framework called Castling-ViT, which trains ViTs using both linear-angular attention and masked softmax-based quadratic attention, but then switches to having only linear angular attention during ViT inference. Our Castling-ViT leverages angular kernels to measure the similarities between queries and keys via spectral angles. And we further simplify it with two techniques: (1) a novel linear-angular attention mechanism: we decompose the angular kernels into linear terms and high-order residuals, and only keep the linear terms; and (2) we adopt two parameterized modules to approximate high-order residuals: a depthwise convolution and an auxiliary masked softmax attention to help learn both global and local information, where the masks for softmax attention are regularized to gradually become zeros and thus incur no overhead during ViT inference. Extensive experiments and ablation studies on three tasks consistently validate the effectiveness of the proposed Castling-ViT, e.g., achieving up to a 1.8% higher accuracy or 40% MACs reduction on ImageNet classification and 1.2 higher mAP on COCO detection under comparable FLOPs, as compared to ViTs with vanilla softmax-based attentions.
Authors: Xiangchao Yan, Runjian Chen, Bo Zhang, Hancheng Ye, Renqiu Xia, Jiakang Yuan, Hongbin Zhou, Xinyu Cai, Botian Shi, Wenqi Shao, Ping Luo, Yu Qiao, Tao Chen, Junchi Yan
Abstract: Annotating 3D LiDAR point clouds for perception tasks is fundamental for many applications e.g., autonomous driving, yet it still remains notoriously labor-intensive. Pretraining-finetuning approach can alleviate the labeling burden by fine-tuning a pre-trained backbone across various downstream datasets as well as tasks. In this paper, we propose SPOT, namely Scalable Pre-training via Occupancy prediction for learning Transferable 3D representations under such a label-efficient fine-tuning paradigm. SPOT achieves effectiveness on various public datasets with different downstream tasks, showcasing its general representation power, cross-domain robustness and data scalability which are three key factors for real-world application. Specifically, we both theoretically and empirically show, for the first time, that general representations learning can be achieved through the task of occupancy prediction. Then, to address the domain gap caused by different LiDAR sensors and annotation methods, we develop a beam re-sampling technique for point cloud augmentation combined with class-balancing strategy. Furthermore, scalable pre-training is observed, that is, the downstream performance across all the experiments gets better with more pre-training data. Additionally, such pre-training strategy also remains compatible with unlabeled data. The hope is that our findings will facilitate the understanding of LiDAR points and pave the way for future advancements in LiDAR pre-training.
Authors: Xizhe Xue, Haokui Zhang, Zongwen Bai, Ying Li
Abstract: Data-hungry HSI classification methods require high-quality labeled HSIs, which are often costly to obtain. This characteristic limits the performance potential of data-driven methods when dealing with limited annotated samples. Bridging the domain gap between data acquired from different sensors allows us to utilize abundant labeled data across sensors to break this bottleneck. In this paper, we propose a novel Attention-Gated Tuning (AGT) strategy and a triplet-structured transformer model, Tri-Former, to address this issue. The AGT strategy serves as a bridge, allowing us to leverage existing labeled HSI datasets, even RGB datasets to enhance the performance on new HSI datasets with limited samples. Instead of inserting additional parameters inside the basic model, we train a lightweight auxiliary branch that takes intermediate features as input from the basic model and makes predictions. The proposed AGT resolves conflicts between heterogeneous and even cross-modal data by suppressing the disturbing information and enhances the useful information through a soft gate. Additionally, we introduce Tri-Former, a triplet-structured transformer with a spectral-spatial separation design that enhances parameter utilization and computational efficiency, enabling easier and flexible fine-tuning. Comparison experiments conducted on three representative HSI datasets captured by different sensors demonstrate the proposed Tri-Former achieves better performance compared to several state-of-the-art methods. Homologous, heterologous and cross-modal tuning experiments verified the effectiveness of the proposed AGT. Code has been released at: \href{https://github.com/Cecilia-xue/AGT}{https://github.com/Cecilia-xue/AGT}.
URLs: https://github.com/Cecilia-xue/AGT, https://github.com/Cecilia-xue/AGT
Authors: Xiaochao Pan, Jiawei Yao, Hongrui Kou, Tong Wu, Canran Xiao
Abstract: In the realm of autonomous driving, achieving precise 3D reconstruction of the driving environment is critical for ensuring safety and effective navigation. Neural Radiance Fields (NeRF) have shown promise in creating highly detailed and accurate models of complex environments. However, the application of NeRF in autonomous driving scenarios encounters several challenges, primarily due to the sparsity of viewpoints inherent in camera trajectories and the constraints on data collection in unbounded outdoor scenes, which typically occur along predetermined paths. This limitation not only reduces the available scene information but also poses significant challenges for NeRF training, as the sparse and path-distributed observational data leads to under-representation of the scene's geometry. In this paper, we introduce HarmonicNeRF, a novel approach for outdoor self-supervised monocular scene reconstruction. HarmonicNeRF capitalizes on the strengths of NeRF and enhances surface reconstruction accuracy by augmenting the input space with geometry-informed synthetic views. This is achieved through the application of spherical harmonics to generate novel radiance values, taking into careful consideration the color observations from the limited available real-world views. Additionally, our method incorporates proxy geometry to effectively manage occlusion, generating radiance pseudo-labels that circumvent the limitations of traditional image-warping techniques, which often fail in sparse data conditions typical of autonomous driving environments. Extensive experiments conducted on the KITTI, Argoverse, and NuScenes datasets demonstrate our approach establishes new benchmarks in synthesizing novel depth views and reconstructing scenes, significantly outperforming existing methods. Project page: https://github.com/Jiawei-Yao0812/HarmonicNeRF
Authors: Jiawei Yao, Yingxin Lai, Hongrui Kou, Tong Wu, Ruixi Liu
Abstract: 3D object detection plays a pivotal role in autonomous driving and robotics, demanding precise interpretation of Bird's Eye View (BEV) images. The dynamic nature of real-world environments necessitates the use of dynamic query mechanisms in 3D object detection to adaptively capture and process the complex spatio-temporal relationships present in these scenes. However, prior implementations of dynamic queries have often faced difficulties in effectively leveraging these relationships, particularly when it comes to integrating temporal information in a computationally efficient manner. Addressing this limitation, we introduce a framework utilizing dynamic query evolution strategy, harnesses K-means clustering and Top-K attention mechanisms for refined spatio-temporal data processing. By dynamically segmenting the BEV space and prioritizing key features through Top-K attention, our model achieves a real-time, focused analysis of pertinent scene elements. Our extensive evaluation on the nuScenes and Waymo dataset showcases a marked improvement in detection accuracy, setting a new benchmark in the domain of query-based BEV object detection. Our dynamic query evolution strategy has the potential to push the boundaries of current BEV methods with enhanced adaptability and computational efficiency. Project page: https://github.com/Jiawei-Yao0812/QE-BEV
Authors: Andreas Reich, Mirko Maehlisch
Abstract: Low-latency instance segmentation of LiDAR point clouds is crucial in real-world applications because it serves as an initial and frequently-used building block in a robot's perception pipeline, where every task adds further delay. Particularly in dynamic environments, this total delay can result in significant positional offsets of dynamic objects, as seen in highway scenarios. To address this issue, we employ a new technique, which we call continuous clustering. Unlike most existing clustering approaches, which use a full revolution of the LiDAR sensor, we process the data stream in a continuous and seamless fashion. Our approach does not rely on the concept of complete or partial sensor rotations with multiple discrete range images; instead, it views the range image as a single and infinitely horizontally growing entity. Each new column of this continuous range image is processed as soon it is available. Obstacle points are clustered to existing instances in real-time and it is checked at a high-frequency which instances are completed in order to publish them without waiting for the completion of the revolution or some other integration period. In the case of rotating sensors, no problematic discontinuities between the points of the end and the start of a scan are observed. In this work we describe the two-layered data structure and the corresponding algorithm for continuous clustering. It is able to achieve an average latency of just 5 ms with respect to the latest timestamp of all points in the cluster. We are publishing the source code at https://github.com/UniBwTAS/continuous_clustering.
Authors: Dongkeun Kim, Youngkil Song, Minsu Cho, Suha Kwak
Abstract: Group activity detection (GAD) is the task of identifying members of each group and classifying the activity of the group at the same time in a video. While GAD has been studied recently, there is still much room for improvement in both dataset and methodology due to their limited capability to address practical GAD scenarios. To resolve these issues, we first present a new dataset, dubbed Caf\'e. Unlike existing datasets, Caf\'e is constructed primarily for GAD and presents more practical scenarios and metrics, as well as being large-scale and providing rich annotations. Along with the dataset, we propose a new GAD model that deals with an unknown number of groups and latent group members efficiently and effectively. We evaluated our model on three datasets including Caf\'e, where it outperformed previous work in terms of both accuracy and inference speed.
Authors: Ashwath Shetty, Marc Habermann, Guoxing Sun, Diogo Luvizon, Vladislav Golyanik, Christian Theobalt
Abstract: We present the first approach to render highly realistic free-viewpoint videos of a human actor in general apparel, from sparse multi-view recording to display, in real-time at an unprecedented 4K resolution. At inference, our method only requires four camera views of the moving actor and the respective 3D skeletal pose. It handles actors in wide clothing, and reproduces even fine-scale dynamic detail, e.g. clothing wrinkles, face expressions, and hand gestures. At training time, our learning-based approach expects dense multi-view video and a rigged static surface scan of the actor. Our method comprises three main stages. Stage 1 is a skeleton-driven neural approach for high-quality capture of the detailed dynamic mesh geometry. Stage 2 is a novel solution to create a view-dependent texture using four test-time camera views as input. Finally, stage 3 comprises a new image-based refinement network rendering the final 4K image given the output from the previous stages. Our approach establishes a new benchmark for real-time rendering resolution and quality using sparse input camera views, unlocking possibilities for immersive telepresence.
Authors: Tianxing Wu, Chenyang Si, Yuming Jiang, Ziqi Huang, Ziwei Liu
Abstract: Though diffusion-based video generation has witnessed rapid progress, the inference results of existing models still exhibit unsatisfactory temporal consistency and unnatural dynamics. In this paper, we delve deep into the noise initialization of video diffusion models, and discover an implicit training-inference gap that attributes to the unsatisfactory inference quality.Our key findings are: 1) the spatial-temporal frequency distribution of the initial noise at inference is intrinsically different from that for training, and 2) the denoising process is significantly influenced by the low-frequency components of the initial noise. Motivated by these observations, we propose a concise yet effective inference sampling strategy, FreeInit, which significantly improves temporal consistency of videos generated by diffusion models. Through iteratively refining the spatial-temporal low-frequency components of the initial latent during inference, FreeInit is able to compensate the initialization gap between training and inference, thus effectively improving the subject appearance and temporal consistency of generation results. Extensive experiments demonstrate that FreeInit consistently enhances the generation quality of various text-to-video diffusion models without additional training or fine-tuning.
Authors: Jaesung Rim, Junyong Lee, Heemin Yang, Sunghyun Cho
Abstract: Mobile cameras, despite their significant advancements, still have difficulty in low-light imaging due to compact sensors and lenses, leading to longer exposures and motion blur. Traditional blind deconvolution methods and learning-based deblurring methods can be potential solutions to remove blur. However, achieving practical performance still remains a challenge. To address this, we propose a learning-based deblurring framework for smartphones, utilizing wide and ultra-wide cameras as a hybrid camera system. We simultaneously capture a long-exposure wide image and short-exposure burst ultra-wide images, and utilize the burst images to deblur the wide image. To fully exploit burst ultra-wide images, we present HCDeblur, a practical deblurring framework that includes novel deblurring networks, HC-DNet and HC-FNet. HC-DNet utilizes motion information extracted from burst images to deblur a wide image, and HC-FNet leverages burst images as reference images to further enhance a deblurred output. For training and evaluating the proposed method, we introduce the HCBlur dataset, which consists of synthetic and real-world datasets. Our experiments demonstrate that HCDeblur achieves state-of-the-art deblurring quality. Code and datasets are available at https://cg.postech.ac.kr/research/HCDeblur.
Authors: A. Emin Orhan, Wentao Wang, Alex N. Wang, Mengye Ren, Brenden M. Lake
Abstract: Children learn powerful internal models of the world around them from a few years of egocentric visual experience. Can such internal models be learned from a child's visual experience with highly generic learning algorithms or do they require strong inductive biases? Recent advances in collecting large-scale, longitudinal, developmentally realistic video datasets and generic self-supervised learning (SSL) algorithms are allowing us to begin to tackle this nature vs. nurture question. However, existing work typically focuses on image-based SSL algorithms and visual capabilities that can be learned from static images (e.g. object recognition), thus ignoring temporal aspects of the world. To close this gap, here we train self-supervised video models on longitudinal, egocentric headcam recordings collected from a child over a two year period in their early development (6-31 months). The resulting models are highly effective at facilitating the learning of action concepts from a small number of labeled examples; they have favorable data size scaling properties; and they display emergent video interpolation capabilities. Video models also learn more robust object representations than image-based models trained with the exact same data. These results suggest that important temporal aspects of a child's internal model of the world may be learnable from their visual experience using highly generic learning algorithms and without strong inductive biases.
Authors: Ifeoma Veronica Nwabufo, Jan Niklas B\"ohm, Philipp Berens, Dmitry Kobak
Abstract: Self-supervised learning methods based on data augmentations, such as SimCLR, BYOL, or DINO, allow obtaining semantically meaningful representations of image datasets and are widely used prior to supervised fine-tuning. A recent self-supervised learning method, $t$-SimCNE, uses contrastive learning to directly train a 2D representation suitable for visualisation. When applied to natural image datasets, $t$-SimCNE yields 2D visualisations with semantically meaningful clusters. In this work, we used $t$-SimCNE to visualise medical image datasets, including examples from dermatology, histology, and blood microscopy. We found that increasing the set of data augmentations to include arbitrary rotations improved the results in terms of class separability, compared to data augmentations used for natural images. Our 2D representations show medically relevant structures and can be used to aid data exploration and annotation, improving on common approaches for data visualisation.
Authors: Seong Hun Lee, Javier Civera, Patrick Vandewalle
Abstract: We propose a robust method for point cloud registration that can handle both unknown scales and extreme outlier ratios. Our method, dubbed PCR-99, uses a deterministic 3-point sampling approach with two novel mechanisms that significantly boost the speed: (1) an improved ordering of the samples based on pairwise scale consistency, prioritizing the point correspondences that are more likely to be inliers, and (2) an efficient outlier rejection scheme based on triplet scale consistency, prescreening bad samples and reducing the number of hypotheses to be tested. Our evaluation shows that, up to 98% outlier ratio, the proposed method achieves comparable performance to the state of the art. At 99% outlier ratio, however, it outperforms the state of the art for both known-scale and unknown-scale problems. Especially for the latter, we observe a clear superiority in terms of robustness and speed.
Authors: Wenhao Tang, Fengtao Zhou, Sheng Huang, Xiang Zhu, Yi Zhang, Bo Liu
Abstract: Multiple instance learning (MIL) is the most widely used framework in computational pathology, encompassing sub-typing, diagnosis, prognosis, and more. However, the existing MIL paradigm typically requires an offline instance feature extractor, such as a pre-trained ResNet or a foundation model. This approach lacks the capability for feature fine-tuning within the specific downstream tasks, limiting its adaptability and performance. To address this issue, we propose a Re-embedded Regional Transformer (R$^2$T) for re-embedding the instance features online, which captures fine-grained local features and establishes connections across different regions. Unlike existing works that focus on pre-training powerful feature extractor or designing sophisticated instance aggregator, R$^2$T is tailored to re-embed instance features online. It serves as a portable module that can seamlessly integrate into mainstream MIL models. Extensive experimental results on common computational pathology tasks validate that: 1) feature re-embedding improves the performance of MIL models based on ResNet-50 features to the level of foundation model features, and further enhances the performance of foundation model features; 2) the R$^2$T can introduce more significant performance improvements to various MIL models; 3) R$^2$T-MIL, as an R$^2$T-enhanced AB-MIL, outperforms other latest methods by a large margin.The code is available at: https://github.com/DearCaat/RRT-MIL.
Authors: Haozhi Cao, Yuecong Xu, Jianfei Yang, Pengyu Yin, Xingyu Ji, Shenghai Yuan, Lihua Xie
Abstract: Multi-modal test-time adaptation (MM-TTA) is proposed to adapt models to an unlabeled target domain by leveraging the complementary multi-modal inputs in an online manner. Previous MM-TTA methods for 3D segmentation rely on predictions of cross-modal information in each input frame, while they ignore the fact that predictions of geometric neighborhoods within consecutive frames are highly correlated, leading to unstable predictions across time. To fulfill this gap, we propose ReLiable Spatial-temporal Voxels (Latte), an MM-TTA method that leverages reliable cross-modal spatial-temporal correspondences for multi-modal 3D segmentation. Motivated by the fact that reliable predictions should be consistent with their spatial-temporal correspondences, Latte aggregates consecutive frames in a slide window manner and constructs Spatial-Temopral (ST) voxels to capture temporally local prediction consistency for each modality. After filtering out ST voxels with high ST entropy, Latte conducts cross-modal learning for each point and pixel by attending to those with reliable and consistent predictions among both spatial and temporal neighborhoods. Experimental results show that Latte achieves state-of-the-art performance on three different MM-TTA benchmarks compared to previous MM-TTA or TTA methods. Visit our project site https://sites.google.com/view/eccv24-latte.
Authors: Jianan Jiang, Xinglin Li, Weiren Yu, Di Wu
Abstract: In the realm of fashion design, sketches serve as the canvas for expressing an artist's distinctive drawing style and creative vision, capturing intricate details like stroke variations and texture nuances. The advent of sketch-to-image cross-modal translation technology has notably aided designers. However, existing methods often compromise these sketch details during image generation, resulting in images that deviate from the designer's intended concept. This limitation hampers the ability to offer designers a precise preview of the final output. To overcome this challenge, we introduce HAIFIT, a novel approach that transforms sketches into high-fidelity, lifelike clothing images by integrating multi-scale features and capturing extensive feature map dependencies from diverse perspectives. Through extensive qualitative and quantitative evaluations conducted on our self-collected dataset, our method demonstrates superior performance compared to existing methods in generating photorealistic clothing images. Our method excels in preserving the distinctive style and intricate details essential for fashion design applications. In addition, our method also has obvious advantages in model training and inference speed, contributing to reducing designers' time costs and improving design efficiency.
Authors: Aljo\v{s}a O\v{s}ep, Tim Meinhardt, Francesco Ferroni, Neehar Peri, Deva Ramanan, Laura Leal-Taix\'e
Abstract: We propose the SAL (Segment Anything in Lidar) method consisting of a text-promptable zero-shot model for segmenting and classifying any object in Lidar, and a pseudo-labeling engine that facilitates model training without manual supervision. While the established paradigm for Lidar Panoptic Segmentation (LPS) relies on manual supervision for a handful of object classes defined a priori, we utilize 2D vision foundation models to generate 3D supervision ``for free''. Our pseudo-labels consist of instance masks and corresponding CLIP tokens, which we lift to Lidar using calibrated multi-modal data. By training our model on these labels, we distill the 2D foundation models into our Lidar SAL model. Even without manual labels, our model achieves $91\%$ in terms of class-agnostic segmentation and $54\%$ in terms of zero-shot Lidar Panoptic Segmentation of the fully supervised state-of-the-art. Furthermore, we outperform several baselines that do not distill but only lift image features to 3D. More importantly, we demonstrate that SAL supports arbitrary class prompts, can be easily extended to new datasets, and shows significant potential to improve with increasing amounts of self-labeled data. Code and models are available at this $\href{https://github.com/nv-dvl/segment-anything-lidar}{URL}$.
Authors: Yi Wang, Kunchang Li, Xinhao Li, Jiashuo Yu, Yinan He, Chenting Wang, Guo Chen, Baoqi Pei, Ziang Yan, Rongkun Zheng, Jilan Xu, Zun Wang, Yansong Shi, Tianxiang Jiang, Songze Li, Hongjie Zhang, Yifei Huang, Yu Qiao, Yali Wang, Limin Wang
Abstract: We introduce InternVideo2, a new family of video foundation models (ViFM) that achieve the state-of-the-art results in video recognition, video-text tasks, and video-centric dialogue. Our core design is a progressive training approach that unifies the masked video modeling, crossmodal contrastive learning, and next token prediction, scaling up the video encoder size to 6B parameters. At the data level, we prioritize spatiotemporal consistency by semantically segmenting videos and generating video-audio-speech captions. This improves the alignment between video and text. Through extensive experiments, we validate our designs and demonstrate superior performance on over 60 video and audio tasks. Notably, our model outperforms others on various video-related dialogue and long video understanding benchmarks, highlighting its ability to reason and comprehend longer contexts. Code and models are available at https://github.com/OpenGVLab/InternVideo/tree/main/InternVideo2/.
URLs: https://github.com/OpenGVLab/InternVideo/tree/main/InternVideo2/.
Authors: Yuhuan Yang, Chaofan Ma, Jiangchao Yao, Zhun Zhong, Ya Zhang, Yanfeng Wang
Abstract: Referring Image Segmentation~(RIS) leveraging transformers has achieved great success on the interpretation of complex visual-language tasks. However, the quadratic computation cost makes it resource-consuming in capturing long-range visual-language dependencies. Fortunately, Mamba addresses this with efficient linear complexity in processing. However, directly applying Mamba to multi-modal interactions presents challenges, primarily due to inadequate channel interactions for the effective fusion of multi-modal data. In this paper, we propose ReMamber, a novel RIS architecture that integrates the power of Mamba with a multi-modal Mamba Twister block. The Mamba Twister explicitly models image-text interaction, and fuses textual and visual features through its unique channel and spatial twisting mechanism. We achieve competitive results on three challenging benchmarks with a simple and efficient architecture. Moreover, we conduct thorough analyses of ReMamber and discuss other fusion designs using Mamba. These provide valuable perspectives for future research. The code has been released at: https://github.com/yyh-rain-song/ReMamber.
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. The code is available at https://github.com/wuqianliang/Diff-Reg.
Authors: Jeongmin Bae, Seoha Kim, Youngsik Yun, Hahyun Lee, Gun Bang, Youngjung Uh
Abstract: As 3D Gaussian Splatting (3DGS) provides fast and high-quality novel view synthesis, it is a natural extension to deform a canonical 3DGS to multiple frames for representing a dynamic scene. However, previous works fail to accurately reconstruct complex dynamic scenes. We attribute the failure to the design of the deformation field, which is built as a coordinate-based function. This approach is problematic because 3DGS is a mixture of multiple fields centered at the Gaussians, not just a single coordinate-based framework. To resolve this problem, we define the deformation as a function of per-Gaussian embeddings and temporal embeddings. Moreover, we decompose deformations as coarse and fine deformations to model slow and fast movements, respectively. Also, we introduce a local smoothness regularization for per-Gaussian embedding to improve the details in dynamic regions. Project page: https://jeongminb.github.io/e-d3dgs/
Authors: Bahri Batuhan Bilecen, Yigit Yalin, Ning Yu, Aysegul Dundar
Abstract: Generative Adversarial Networks (GANs) have emerged as powerful tools for high-quality image generation and real image editing by manipulating their latent spaces. Recent advancements in GANs include 3D-aware models such as EG3D, which feature efficient triplane-based architectures capable of reconstructing 3D geometry from single images. However, limited attention has been given to providing an integrated framework for 3D-aware, high-quality, reference-based image editing. This study addresses this gap by exploring and demonstrating the effectiveness of the triplane space for advanced reference-based edits. Our novel approach integrates encoding, automatic localization, spatial disentanglement of triplane features, and fusion learning to achieve the desired edits. Additionally, our framework demonstrates versatility and robustness across various domains, extending its effectiveness to animal face edits, partially stylized edits like cartoon faces, full-body clothing edits, and 360-degree head edits. Our method shows state-of-the-art performance over relevant latent direction, text, and image-guided 2D and 3D-aware diffusion and GAN methods, both qualitatively and quantitatively.
Authors: Shijie Zhou, Zhiwen Fan, Dejia Xu, Haoran Chang, Pradyumna Chari, Tejas Bharadwaj, Suya You, Zhangyang Wang, Achuta Kadambi
Abstract: The increasing demand for virtual reality applications has highlighted the significance of crafting immersive 3D assets. We present a text-to-3D 360$^{\circ}$ scene generation pipeline that facilitates the creation of comprehensive 360$^{\circ}$ scenes for in-the-wild environments in a matter of minutes. Our approach utilizes the generative power of a 2D diffusion model and prompt self-refinement to create a high-quality and globally coherent panoramic image. This image acts as a preliminary "flat" (2D) scene representation. Subsequently, it is lifted into 3D Gaussians, employing splatting techniques to enable real-time exploration. To produce consistent 3D geometry, our pipeline constructs a spatially coherent structure by aligning the 2D monocular depth into a globally optimized point cloud. This point cloud serves as the initial state for the centroids of 3D Gaussians. In order to address invisible issues inherent in single-view inputs, we impose semantic and geometric constraints on both synthesized and input camera views as regularizations. These guide the optimization of Gaussians, aiding in the reconstruction of unseen regions. In summary, our method offers a globally consistent 3D scene within a 360$^{\circ}$ perspective, providing an enhanced immersive experience over existing techniques. Project website at: http://dreamscene360.github.io/
Authors: Fabian Isensee, Tassilo Wald, Constantin Ulrich, Michael Baumgartner, Saikat Roy, Klaus Maier-Hein, Paul F. Jaeger
Abstract: The release of nnU-Net marked a paradigm shift in 3D medical image segmentation, demonstrating that a properly configured U-Net architecture could still achieve state-of-the-art results. Despite this, the pursuit of novel architectures, and the respective claims of superior performance over the U-Net baseline, continued. In this study, we demonstrate that many of these recent claims fail to hold up when scrutinized for common validation shortcomings, such as the use of inadequate baselines, insufficient datasets, and neglected computational resources. By meticulously avoiding these pitfalls, we conduct a thorough and comprehensive benchmarking of current segmentation methods including CNN-based, Transformer-based, and Mamba-based approaches. In contrast to current beliefs, we find that the recipe for state-of-the-art performance is 1) employing CNN-based U-Net models, including ResNet and ConvNeXt variants, 2) using the nnU-Net framework, and 3) scaling models to modern hardware resources. These results indicate an ongoing innovation bias towards novel architectures in the field and underscore the need for more stringent validation standards in the quest for scientific progress.
Authors: Dmytro Shvetsov, Joonas Ariva, Marharyta Domnich, Raul Vicente, Dmytro Fishman
Abstract: Deep learning is dramatically transforming the field of medical imaging and radiology, enabling the identification of pathologies in medical images, including computed tomography (CT) and X-ray scans. However, the performance of deep learning models, particularly in segmentation tasks, is often limited by the need for extensive annotated datasets. To address this challenge, the capabilities of weakly supervised semantic segmentation are explored through the lens of Explainable AI and the generation of counterfactual explanations. The scope of this research is development of a novel counterfactual inpainting approach (COIN) that flips the predicted classification label from abnormal to normal by using a generative model. For instance, if the classifier deems an input medical image X as abnormal, indicating the presence of a pathology, the generative model aims to inpaint the abnormal region, thus reversing the classifier's original prediction label. The approach enables us to produce precise segmentations for pathologies without depending on pre-existing segmentation masks. Crucially, image-level labels are utilized, which are substantially easier to acquire than creating detailed segmentation masks. The effectiveness of the method is demonstrated by segmenting synthetic targets and actual kidney tumors from CT images acquired from Tartu University Hospital in Estonia. The findings indicate that COIN greatly surpasses established attribution methods, such as RISE, ScoreCAM, and LayerCAM, as well as an alternative counterfactual explanation method introduced by Singla et al. This evidence suggests that COIN is a promising approach for semantic segmentation of tumors in CT images, and presents a step forward in making deep learning applications more accessible and effective in healthcare, where annotated data is scarce.
Authors: Trung Dang, Huy Hoang Nguyen, Aleksei Tiulpin
Abstract: One of the primary challenges in brain tumor segmentation arises from the uncertainty of voxels close to tumor boundaries. However, the conventional process of generating ground truth segmentation masks fails to treat such uncertainties properly. Those "hard labels" with 0s and 1s conceptually influenced the majority of prior studies on brain image segmentation. As a result, tumor segmentation is often solved through voxel classification. In this work, we instead view this problem as a voxel-level regression, where the ground truth represents a certainty mapping from any pixel to the border of the tumor. We propose a novel ground truth label transformation, which is based on a signed geodesic transform, to capture the uncertainty in brain tumors' vicinity. We combine this idea with a Focal-like regression L1-loss that enables effective regression learning in high-dimensional output space by appropriately weighting voxels according to their difficulty. We thoroughly conduct an experimental evaluation to validate the components of our proposed method, compare it to a diverse array of state-of-the-art segmentation models, and show that it is architecture-agnostic. The code of our method is made publicly available (\url{https://github.com/Oulu-IMEDS/SiNGR/}).
Authors: Trung Dang, Huy Hoang Nguyen, Aleksei Tiulpin
Abstract: Accurate retinal vessel (RV) segmentation is a crucial step in the quantitative assessment of retinal vasculature, which is needed for the early detection of retinal diseases and other conditions. Numerous studies have been conducted to tackle the problem of segmenting vessels automatically using a pixel-wise classification approach. The common practice of creating ground truth labels is to categorize pixels as foreground and background. This approach is, however, biased, and it ignores the uncertainty of a human annotator when it comes to annotating e.g. thin vessels. In this work, we propose a simple and effective method that casts the RV segmentation task as an image-level regression. For this purpose, we first introduce a novel Segmentation Annotation Uncertainty-Aware (SAUNA) transform, which adds pixel uncertainty to the ground truth using the pixel's closeness to the annotation boundary and vessel thickness. To train our model with soft labels, we generalize the earlier proposed Jaccard metric loss to arbitrary hypercubes for soft Jaccard index (Intersection-over-Union) optimization. Additionally, we employ a stable version of the Focal-L1 loss for pixel-wise regression. We conduct thorough experiments and compare our method to a diverse set of baselines across 5 retinal image datasets. Our empirical results indicate that the integration of the SAUNA transform and these segmentation losses led to significant performance boosts for different segmentation models. Particularly, our methodology enables UNet-like architectures to substantially outperform computational-intensive baselines. Our implementation is available at \url{https://github.com/Oulu-IMEDS/SAUNA}.
Authors: Saehyung Lee, Sangwon Yu, Junsung Park, Jihun Yi, Sungroh Yoon
Abstract: In this paper, we primarily address the issue of dialogue-form context query within the interactive text-to-image retrieval task. Our methodology, PlugIR, actively utilizes the general instruction-following capability of LLMs in two ways. First, by reformulating the dialogue-form context, we eliminate the necessity of fine-tuning a retrieval model on existing visual dialogue data, thereby enabling the use of any arbitrary black-box model. Second, we construct the LLM questioner to generate non-redundant questions about the attributes of the target image, based on the information of retrieval candidate images in the current context. This approach mitigates the issues of noisiness and redundancy in the generated questions. Beyond our methodology, we propose a novel evaluation metric, Best log Rank Integral (BRI), for a comprehensive assessment of the interactive retrieval system. PlugIR demonstrates superior performance compared to both zero-shot and fine-tuned baselines in various benchmarks. Additionally, the two methodologies comprising PlugIR can be flexibly applied together or separately in various situations. Our codes are available at https://github.com/Saehyung-Lee/PlugIR.
Authors: Utsab Saha, Sawradip Saha, Shaikh Anowarul Fattah, Mohammad Saquib
Abstract: The identification and restoration of ancient watermarks have long been a major topic in codicology and history. Classifying historical documents based on watermarks is challenging due to their diversity, noisy samples, multiple representation modes, and minor distinctions between classes and intra-class variations. This paper proposes a modified U-net-based conditional generative adversarial network (GAN) named Npix2Cpix to translate noisy raw historical watermarked images into clean, handwriting-free watermarked images by performing image translation from degraded (noisy) pixels to clean pixels. Using image-to-image translation and adversarial learning, the network creates clutter-free images for watermark restoration and categorization. The generator and discriminator of the proposed GAN are trained using two separate loss functions, each based on the distance between images, to learn the mapping from the input noisy image to the output clean image. After using the proposed GAN to pre-process noisy watermarked images, Siamese-based one-shot learning is employed for watermark classification. Experimental results on a large-scale historical watermark dataset demonstrate that cleaning the noisy watermarked images can help to achieve high one-shot classification accuracy. The qualitative and quantitative evaluation of the retrieved watermarked image highlights the effectiveness of the proposed approach.
Authors: Weijie Wei, Osman \"Ulger, Fatemeh Karimi Nejadasl, Theo Gevers, Martin R. Oswald
Abstract: Existing perception methods for autonomous driving fall short of recognizing unknown entities not covered in the training data. Open-vocabulary methods offer promising capabilities in detecting any object but are limited by user-specified queries representing target classes. We propose AutoVoc3D, a framework for automatic object class recognition and open-ended segmentation. Evaluation on nuScenes showcases AutoVoc3D's ability to generate precise semantic classes and accurate point-wise segmentation. Moreover, we introduce Text-Point Semantic Similarity, a new metric to assess the semantic similarity between text and point cloud without eliminating novel classes.
Authors: Changan Chen, Puyuan Peng, Ami Baid, Zihui Xue, Wei-Ning Hsu, David Harwath, Kristen Grauman
Abstract: Generating realistic audio for human actions is important for many applications, such as creating sound effects for films or virtual reality games. Existing approaches implicitly assume total correspondence between the video and audio during training, yet many sounds happen off-screen and have weak to no correspondence with the visuals -- resulting in uncontrolled ambient sounds or hallucinations at test time. We propose a novel ambient-aware audio generation model, AV-LDM. We devise a novel audio-conditioning mechanism to learn to disentangle foreground action sounds from the ambient background sounds in in-the-wild training videos. Given a novel silent video, our model uses retrieval-augmented generation to create audio that matches the visual content both semantically and temporally. We train and evaluate our model on two in-the-wild egocentric video datasets, Ego4D and EPIC-KITCHENS, and we introduce Ego4D-Sounds -- 1.2M curated clips with action-audio correspondence. Our model outperforms an array of existing methods, allows controllable generation of the ambient sound, and even shows promise for generalizing to computer graphics game clips. Overall, our approach is the first to focus video-to-audio generation faithfully on the observed visual content despite training from uncurated clips with natural background sounds.
Authors: Sunny Duan, Mikail Khona, Abhiram Iyer, Rylan Schaeffer, Ila R Fiete
Abstract: Frontier AI systems are making transformative impacts across society, but such benefits are not without costs: models trained on web-scale datasets containing personal and private data raise profound concerns about data privacy and security. Language models are trained on extensive corpora including potentially sensitive or proprietary information, and the risk of data leakage - where the model response reveals pieces of such information - remains inadequately understood. Prior work has investigated what factors drive memorization and have identified that sequence complexity and the number of repetitions drive memorization. Here, we focus on the evolution of memorization over training. We begin by reproducing findings that the probability of memorizing a sequence scales logarithmically with the number of times it is present in the data. We next show that sequences which are apparently not memorized after the first encounter can be "uncovered" throughout the course of training even without subsequent encounters, a phenomenon we term "latent memorization". The presence of latent memorization presents a challenge for data privacy as memorized sequences may be hidden at the final checkpoint of the model but remain easily recoverable. To this end, we develop a diagnostic test relying on the cross entropy loss to uncover latent memorized sequences with high accuracy.
Authors: Ranjan Sapkota, Rizwan Qureshi, Marco Flores Calero, Chetan Badjugar, Upesh Nepal, Alwin Poulose, Peter Zeno, Uday Bhanu Prakash Vaddevolu, Sheheryar Khan, Maged Shoman, Hong Yan, Manoj Karkee
Abstract: This review systematically examines the progression of the You Only Look Once (YOLO) object detection algorithms from YOLOv1 to the recently unveiled YOLOv10. Employing a reverse chronological analysis, this study examines the advancements introduced by YOLO algorithms, beginning with YOLOv10 and progressing through YOLOv9, YOLOv8, and subsequent versions to explore each version's contributions to enhancing speed, accuracy, and computational efficiency in real-time object detection. The study highlights the transformative impact of YOLO across five critical application areas: automotive safety, healthcare, industrial manufacturing, surveillance, and agriculture. By detailing the incremental technological advancements in subsequent YOLO versions, this review chronicles the evolution of YOLO, and discusses the challenges and limitations in each earlier versions. The evolution signifies a path towards integrating YOLO with multimodal, context-aware, and General Artificial Intelligence (AGI) systems for the next YOLO decade, promising significant implications for future developments in AI-driven applications.
Authors: Junhao Su, Changpeng Cai, Feiyu Zhu, Chenghao He, Xiaojie Xu, Dongzhi Guan, Chenyang Si
Abstract: Deep neural networks conventionally employ end-to-end backpropagation for their training process, which lacks biological credibility and triggers a locking dilemma during network parameter updates, leading to significant GPU memory use. Supervised local learning, which segments the network into multiple local blocks updated by independent auxiliary networks. However, these methods cannot replace end-to-end training due to lower accuracy, as gradients only propagate within their local block, creating a lack of information exchange between blocks. To address this issue and establish information transfer across blocks, we propose a Momentum Auxiliary Network (MAN) that establishes a dynamic interaction mechanism. The MAN leverages an exponential moving average (EMA) of the parameters from adjacent local blocks to enhance information flow. This auxiliary network, updated through EMA, helps bridge the informational gap between blocks. Nevertheless, we observe that directly applying EMA parameters has certain limitations due to feature discrepancies among local blocks. To overcome this, we introduce learnable biases, further boosting performance. We have validated our method on four image classification datasets (CIFAR-10, STL-10, SVHN, ImageNet), attaining superior performance and substantial memory savings. Notably, our method can reduce GPU memory usage by more than 45\% on the ImageNet dataset compared to end-to-end training, while achieving higher performance. The Momentum Auxiliary Network thus offers a new perspective for supervised local learning. Our code is available at: https://github.com/JunhaoSu0/MAN.
Authors: Nithish Kannen, Arif Ahmad, Marco Andreetto, Vinodkumar Prabhakaran, Utsav Prabhu, Adji Bousso Dieng, Pushpak Bhattacharyya, Shachi Dave
Abstract: Text-to-Image (T2I) models are being increasingly adopted in diverse global communities where they create visual representations of their unique cultures. Current T2I benchmarks primarily focus on faithfulness, aesthetics, and realism of generated images, overlooking the critical dimension of cultural competence. In this work, we introduce a framework to evaluate cultural competence of T2I models along two crucial dimensions: cultural awareness and cultural diversity, and present a scalable approach using a combination of structured knowledge bases and large language models to build a large dataset of cultural artifacts to enable this evaluation. In particular, we apply this approach to build CUBE (CUltural BEnchmark for Text-to-Image models), a first-of-its-kind benchmark to evaluate cultural competence of T2I models. CUBE covers cultural artifacts associated with 8 countries across different geo-cultural regions and along 3 concepts: cuisine, landmarks, and art. CUBE consists of 1) CUBE-1K, a set of high-quality prompts that enable the evaluation of cultural awareness, and 2) CUBE-CSpace, a larger dataset of cultural artifacts that serves as grounding to evaluate cultural diversity. We also introduce cultural diversity as a novel T2I evaluation component, leveraging quality-weighted Vendi score. Our evaluations reveal significant gaps in the cultural awareness of existing models across countries and provide valuable insights into the cultural diversity of T2I outputs for under-specified prompts. Our methodology is extendable to other cultural regions and concepts, and can facilitate the development of T2I models that better cater to the global population.
Authors: Feixiang Zhou, Bryan Williams, Hossein Rahmani
Abstract: Alleviating noisy pseudo labels remains a key challenge in Semi-Supervised Temporal Action Localization (SS-TAL). Existing methods often filter pseudo labels based on strict conditions, but they typically assess classification and localization quality separately, leading to suboptimal pseudo-label ranking and selection. In particular, there might be inaccurate pseudo labels within selected positives, alongside reliable counterparts erroneously assigned to negatives. To tackle these problems, we propose a novel Adaptive Pseudo-label Learning (APL) framework to facilitate better pseudo-label selection. Specifically, to improve the ranking quality, Adaptive Label Quality Assessment (ALQA) is proposed to jointly learn classification confidence and localization reliability, followed by dynamically selecting pseudo labels based on the joint score. Additionally, we propose an Instance-level Consistency Discriminator (ICD) for eliminating ambiguous positives and mining potential positives simultaneously based on inter-instance intrinsic consistency, thereby leading to a more precise selection. We further introduce a general unsupervised Action-aware Contrastive Pre-training (ACP) to enhance the discrimination both within actions and between actions and backgrounds, which benefits SS-TAL. Extensive experiments on THUMOS14 and ActivityNet v1.3 demonstrate that our method achieves state-of-the-art performance under various semi-supervised settings.
Authors: Steven Abreu, Tiffany D. Do, Karan Ahuja, Eric J. Gonzalez, Lee Payne, Daniel McDuff, Mar Gonzalez-Franco
Abstract: Intelligent assistance involves not only understanding but also action. Existing ego-centric video datasets contain rich annotations of the videos, but not of actions that an intelligent assistant could perform in the moment. To address this gap, we release PARSE-Ego4D, a new set of personal action recommendation annotations for the Ego4D dataset. We take a multi-stage approach to generating and evaluating these annotations. First, we used a prompt-engineered large language model (LLM) to generate context-aware action suggestions and identified over 18,000 action suggestions. While these synthetic action suggestions are valuable, the inherent limitations of LLMs necessitate human evaluation. To ensure high-quality and user-centered recommendations, we conducted a large-scale human annotation study that provides grounding in human preferences for all of PARSE-Ego4D. We analyze the inter-rater agreement and evaluate subjective preferences of participants. Based on our synthetic dataset and complete human annotations, we propose several new tasks for action suggestions based on ego-centric videos. We encourage novel solutions that improve latency and energy requirements. The annotations in PARSE-Ego4D will support researchers and developers who are working on building action recommendation systems for augmented and virtual reality systems.
Authors: Shangquan Sun, Wenqi Ren, Xinwei Gao, Rui Wang, Xiaochun Cao
Abstract: Transformer-based image restoration methods in adverse weather have achieved significant progress. Most of them use self-attention along the channel dimension or within spatially fixed-range blocks to reduce computational load. However, such a compromise results in limitations in capturing long-range spatial features. Inspired by the observation that the weather-induced degradation factors mainly cause similar occlusion and brightness, in this work, we propose an efficient Histogram Transformer (Histoformer) for restoring images affected by adverse weather. It is powered by a mechanism dubbed histogram self-attention, which sorts and segments spatial features into intensity-based bins. Self-attention is then applied across bins or within each bin to selectively focus on spatial features of dynamic range and process similar degraded pixels of the long range together. To boost histogram self-attention, we present a dynamic-range convolution enabling conventional convolution to conduct operation over similar pixels rather than neighbor pixels. We also observe that the common pixel-wise losses neglect linear association and correlation between output and ground-truth. Thus, we propose to leverage the Pearson correlation coefficient as a loss function to enforce the recovered pixels following the identical order as ground-truth. Extensive experiments demonstrate the efficacy and superiority of our proposed method. We have released the codes in Github.
Authors: Sunny Gupta, Amit Sethi
Abstract: Federated Learning (FL) offers a privacy-preserving approach to train models on decentralized data. Its potential in healthcare is significant, but challenges arise due to cross-client variations in medical image data, exacerbated by limited annotations. This paper introduces Cross-Client Variations Adaptive Federated Learning (CCVA-FL) to address these issues. CCVA-FL aims to minimize cross-client variations by transforming images into a common feature space. It involves expert annotation of a subset of images from each client, followed by the selection of a client with the least data complexity as the target. Synthetic medical images are then generated using Scalable Diffusion Models with Transformers (DiT) based on the target client's annotated images. These synthetic images, capturing diversity and representing the original data, are shared with other clients. Each client then translates its local images into the target image space using image-to-image translation. The translated images are subsequently used in a federated learning setting to develop a server model. Our results demonstrate that CCVA-FL outperforms Vanilla Federated Averaging by effectively addressing data distribution differences across clients without compromising privacy.
Authors: Jiahao Zhang, Frederic Z. Zhang, Cristian Rodriguez, Yizhak Ben-Shabat, Anoop Cherian, Stephen Gould
Abstract: We study the challenging problem of simultaneously localizing a sequence of queries in the form of instructional diagrams in a video. This requires understanding not only the individual queries but also their interrelationships. However, most existing methods focus on grounding one query at a time, ignoring the inherent structures among queries such as the general mutual exclusiveness and the temporal order. Consequently, the predicted timespans of different step diagrams may overlap considerably or violate the temporal order, thus harming the accuracy. In this paper, we tackle this issue by simultaneously grounding a sequence of step diagrams. Specifically, we propose composite queries, constructed by exhaustively pairing up the visual content features of the step diagrams and a fixed number of learnable positional embeddings. Our insight is that self-attention among composite queries carrying different content features suppress each other to reduce timespan overlaps in predictions, while the cross-attention corrects the temporal misalignment via content and position joint guidance. We demonstrate the effectiveness of our approach on the IAW dataset for grounding step diagrams and the YouCook2 benchmark for grounding natural language queries, significantly outperforming existing methods while simultaneously grounding multiple queries.
Authors: Mohammad Erfan Sadeghi, Arash Fayyazi, Suhas Somashekar, Massoud Pedram
Abstract: Vision Transformers (ViTs) represent a groundbreaking shift in machine learning approaches to computer vision. Unlike traditional approaches, ViTs employ the self-attention mechanism, which has been widely used in natural language processing, to analyze image patches. Despite their advantages in modeling visual tasks, deploying ViTs on hardware platforms, notably Field-Programmable Gate Arrays (FPGAs), introduces considerable challenges. These challenges stem primarily from the non-linear calculations and high computational and memory demands of ViTs. This paper introduces CHOSEN, a software-hardware co-design framework to address these challenges and offer an automated framework for ViT deployment on the FPGAs in order to maximize performance. Our framework is built upon three fundamental contributions: multi-kernel design to maximize the bandwidth, mainly targeting benefits of multi DDR memory banks, approximate non-linear functions that exhibit minimal accuracy degradation, and efficient use of available logic blocks on the FPGA, and efficient compiler to maximize the performance and memory-efficiency of the computing kernels by presenting a novel algorithm for design space exploration to find optimal hardware configuration that achieves optimal throughput and latency. Compared to the state-of-the-art ViT accelerators, CHOSEN achieves a 1.5x and 1.42x improvement in the throughput on the DeiT-S and DeiT-B models.
Authors: Ziming Zhong, Yanxu Xu, Jing Li, Jiale Xu, Zhengxin Li, Chaohui Yu, Shenghua Gao
Abstract: We present MeshSegmenter, a simple yet effective framework designed for zero-shot 3D semantic segmentation. This model successfully extends the powerful capabilities of 2D segmentation models to 3D meshes, delivering accurate 3D segmentation across diverse meshes and segment descriptions. Specifically, our model leverages the Segment Anything Model (SAM) model to segment the target regions from images rendered from the 3D shape. In light of the importance of the texture for segmentation, we also leverage the pretrained stable diffusion model to generate images with textures from 3D shape, and leverage SAM to segment the target regions from images with textures. Textures supplement the shape for segmentation and facilitate accurate 3D segmentation even in geometrically non-prominent areas, such as segmenting a car door within a car mesh. To achieve the 3D segments, we render 2D images from different views and conduct segmentation for both textured and untextured images. Lastly, we develop a multi-view revoting scheme that integrates 2D segmentation results and confidence scores from various views onto the 3D mesh, ensuring the 3D consistency of segmentation results and eliminating inaccuracies from specific perspectives. Through these innovations, MeshSegmenter offers stable and reliable 3D segmentation results both quantitatively and qualitatively, highlighting its potential as a transformative tool in the field of 3D zero-shot segmentation. The code is available at \url{https://github.com/zimingzhong/MeshSegmenter}.
Authors: Boyang Deng, Richard Tucker, Zhengqi Li, Leonidas Guibas, Noah Snavely, Gordon Wetzstein
Abstract: We present a method for generating Streetscapes-long sequences of views through an on-the-fly synthesized city-scale scene. Our generation is conditioned by language input (e.g., city name, weather), as well as an underlying map/layout hosting the desired trajectory. Compared to recent models for video generation or 3D view synthesis, our method can scale to much longer-range camera trajectories, spanning several city blocks, while maintaining visual quality and consistency. To achieve this goal, we build on recent work on video diffusion, used within an autoregressive framework that can easily scale to long sequences. In particular, we introduce a new temporal imputation method that prevents our autoregressive approach from drifting from the distribution of realistic city imagery. We train our Streetscapes system on a compelling source of data-posed imagery from Google Street View, along with contextual map data-which allows users to generate city views conditioned on any desired city layout, with controllable camera poses. Please see more results at our project page at https://boyangdeng.com/streetscapes.
Authors: Bo Yuan, Danpei Zhao, Zhuoran Liu, Wentao Li, Tian Li
Abstract: Continual learning (CL) breaks off the one-way training manner and enables a model to adapt to new data, semantics and tasks continuously. However, current CL methods mainly focus on single tasks. Besides, CL models are plagued by catastrophic forgetting and semantic drift since the lack of old data, which often occurs in remote-sensing interpretation due to the intricate fine-grained semantics. In this paper, we propose Continual Panoptic Perception (CPP), a unified continual learning model that leverages multi-task joint learning covering pixel-level classification, instance-level segmentation and image-level perception for universal interpretation in remote sensing images. Concretely, we propose a collaborative cross-modal encoder (CCE) to extract the input image features, which supports pixel classification and caption generation synchronously. To inherit the knowledge from the old model without exemplar memory, we propose a task-interactive knowledge distillation (TKD) method, which leverages cross-modal optimization and task-asymmetric pseudo-labeling (TPL) to alleviate catastrophic forgetting. Furthermore, we also propose a joint optimization mechanism to achieve end-to-end multi-modal panoptic perception. Experimental results on the fine-grained panoptic perception dataset validate the effectiveness of the proposed model, and also prove that joint optimization can boost sub-task CL efficiency with over 13\% relative improvement on panoptic quality.
Authors: Jaehyeong Jeon, Kibum Kim, Kanghoon Yoon, Chanyoung Park
Abstract: The scene graph generation (SGG) task involves detecting objects within an image and predicting predicates that represent the relationships between the objects. However, in SGG benchmark datasets, each subject-object pair is annotated with a single predicate even though a single predicate may exhibit diverse semantics (i.e., semantic diversity), existing SGG models are trained to predict the one and only predicate for each pair. This in turn results in the SGG models to overlook the semantic diversity that may exist in a predicate, thus leading to biased predictions. In this paper, we propose a novel model-agnostic Semantic Diversity-aware Prototype-based Learning (DPL) framework that enables unbiased predictions based on the understanding of the semantic diversity of predicates. Specifically, DPL learns the regions in the semantic space covered by each predicate to distinguish among the various different semantics that a single predicate can represent. Extensive experiments demonstrate that our proposed model-agnostic DPL framework brings significant performance improvement on existing SGG models, and also effectively understands the semantic diversity of predicates.
Authors: Jinfu Liu, Chen Chen, Mengyuan Liu
Abstract: Skeleton-based action recognition has garnered significant attention due to the utilization of concise and resilient skeletons. Nevertheless, the absence of detailed body information in skeletons restricts performance, while other multimodal methods require substantial inference resources and are inefficient when using multimodal data during both training and inference stages. To address this and fully harness the complementary multimodal features, we propose a novel multi-modality co-learning (MMCL) framework by leveraging the multimodal large language models (LLMs) as auxiliary networks for efficient skeleton-based action recognition, which engages in multi-modality co-learning during the training stage and keeps efficiency by employing only concise skeletons in inference. Our MMCL framework primarily consists of two modules. First, the Feature Alignment Module (FAM) extracts rich RGB features from video frames and aligns them with global skeleton features via contrastive learning. Second, the Feature Refinement Module (FRM) uses RGB images with temporal information and text instruction to generate instructive features based on the powerful generalization of multimodal LLMs. These instructive text features will further refine the classification scores and the refined scores will enhance the model's robustness and generalization in a manner similar to soft labels. Extensive experiments on NTU RGB+D, NTU RGB+D 120 and Northwestern-UCLA benchmarks consistently verify the effectiveness of our MMCL, which outperforms the existing skeleton-based action recognition methods. Meanwhile, experiments on UTD-MHAD and SYSU-Action datasets demonstrate the commendable generalization of our MMCL in zero-shot and domain-adaptive action recognition. Our code is publicly available at: https://github.com/liujf69/MMCL-Action.
Authors: Jingchen Sun, Rohan Sharma, Vishnu Suresh Lokhande, Changyou Chen
Abstract: Prompt Tuning has emerged as a prominent research paradigm for adapting vision-language models to various downstream tasks. However, recent research indicates that prompt tuning methods often lead to overfitting due to limited training samples. In this paper, we propose a Cross-modal Aligned Feature Tuning (Craft) method to address this issue. Cross-modal alignment is conducted by first selecting anchors from the alternative domain and deriving relative representations of the embeddings for the selected anchors. Optimizing for a feature alignment loss over anchor-aligned text and image modalities creates a more unified text-image common space. Overfitting in prompt tuning also deteriorates model performance on out-of-distribution samples. To further improve the prompt model's robustness, we propose minimizing Maximum Mean Discrepancy (MMD) over the anchor-aligned feature spaces to mitigate domain shift. The experiment on four different prompt tuning structures consistently shows the improvement of our method, with increases of up to $6.1\%$ in the Base-to-Novel generalization task, $5.8\%$ in the group robustness task, and $2.7\%$ in the out-of-distribution tasks. The code will be available at https://github.com/Jingchensun/Craft
Authors: Qinfeng Zhu, Ningxin Weng, Lei Fan, Yuanzhi Cai
Abstract: Environmental monitoring of lakeside green areas is crucial for environmental protection. Compared to manual inspections, computer vision technologies offer a more efficient solution when deployed on-site. Multispectral imaging provides diverse information about objects under different spectrums, aiding in the differentiation between waste and lakeside lawn environments. This study introduces WasteMS, the first multispectral dataset established for the semantic segmentation of lakeside waste. WasteMS includes a diverse range of waste types in lawn environments, captured under various lighting conditions. We implemented a rigorous annotation process to label waste in images. Representative semantic segmentation frameworks were used to evaluate segmentation accuracy using WasteMS. Challenges encountered when using WasteMS for segmenting waste on lakeside lawns were discussed. The WasteMS dataset is available at https://github.com/zhuqinfeng1999/WasteMS.
Authors: Preeti Mehta, Aman Sagar, Suchi Kumari
Abstract: An increasing number of classification approaches have been developed to address the issue of image rebroadcast and recapturing, a standard attack strategy in insurance frauds, face spoofing, and video piracy. However, most of them neglected scale variations and domain generalization scenarios, performing poorly in instances involving domain shifts, typically made worse by inter-domain and cross-domain scale variances. To overcome these issues, we propose a cascaded data augmentation and SWIN transformer domain generalization framework (DAST-DG) in the current research work Initially, we examine the disparity in dataset representation. A feature generator is trained to make authentic images from various domains indistinguishable. This process is then applied to recaptured images, creating a dual adversarial learning setup. Extensive experiments demonstrate that our approach is practical and surpasses state-of-the-art methods across different databases. Our model achieves an accuracy of approximately 82\% with a precision of 95\% on high-variance datasets.
Authors: Wanggong Yang, Xiaona Wang, Yingrui Qiu, Yifei Zhao
Abstract: Generating landscape paintings expands the possibilities of artistic creativity and imagination. Traditional landscape painting methods involve using ink or colored ink on rice paper, which requires substantial time and effort. These methods are susceptible to errors and inconsistencies and lack precise control over lines and colors. This paper presents LPGen, a high-fidelity, controllable model for landscape painting generation, introducing a novel multi-modal framework that integrates image prompts into the diffusion model. We extract its edges and contours by computing canny edges from the target landscape image. These, along with natural language text prompts and drawing style references, are fed into the latent diffusion model as conditions. We implement a decoupled cross-attention strategy to ensure compatibility between image and text prompts, facilitating multi-modal image generation. A decoder generates the final image. Quantitative and qualitative analyses demonstrate that our method outperforms existing approaches in landscape painting generation and exceeds the current state-of-the-art. The LPGen network effectively controls the composition and color of landscape paintings, generates more accurate images, and supports further research in deep learning-based landscape painting generation.
Authors: Simon Boeder, Fabian Gigengack, Benjamin Risse
Abstract: The 3D occupancy estimation task has become an important challenge in the area of vision-based autonomous driving recently. However, most existing camera-based methods rely on costly 3D voxel labels or LiDAR scans for training, limiting their practicality and scalability. Moreover, most methods are tied to a predefined set of classes which they can detect. In this work we present a novel approach for open vocabulary occupancy estimation called LangOcc, that is trained only via camera images, and can detect arbitrary semantics via vision-language alignment. In particular, we distill the knowledge of the strong vision-language aligned encoder CLIP into a 3D occupancy model via differentiable volume rendering. Our model estimates vision-language aligned features in a 3D voxel grid using only images. It is trained in a self-supervised manner by rendering our estimations back to 2D space, where ground-truth features can be computed. This training mechanism automatically supervises the scene geometry, allowing for a straight-forward and powerful training method without any explicit geometry supervision. LangOcc outperforms LiDAR-supervised competitors in open vocabulary occupancy by a large margin, solely relying on vision-based training. We also achieve state-of-the-art results in self-supervised semantic occupancy estimation on the Occ3D-nuScenes dataset, despite not being limited to a specific set of categories, thus demonstrating the effectiveness of our proposed vision-language training.
Authors: Hideaki Okamoto, Takakiyo Nomura, Kazuhito Nabeshima, Jun Hashimoto, Hitoshi Iyatomi
Abstract: Endoscopy is widely used to diagnose gastric cancer and has a high diagnostic performance, but because it must be performed by a physician, the number of people who can be diagnosed is limited. Gastric X-ray, on the other hand, can be performed by technicians and can screen a much larger number of patients than endoscopy, but its correct diagnosis requires experience. We propose an unprecedented and practical gastric cancer diagnosis support system for gastric X-ray images, which will enable more people to be screened. The system is based on a general deep learning-based object detection model and includes two novel technical proposals: refined probabilistic stomach image augmentation (R-sGAIA) and hard boundary box learning (HBBT). R-sGAIA is a probabilistic gastric fold region enhancement method that provides more learning patterns for cancer detection models. HBBT is an efficient training method for object detection models that allows the use of unannotated negative (i.e., healthy control) samples that cannot be used for training in conventional detection models, thereby improving model performance. The sensitivity (SE) of the proposed system for gastric cancer (90.2%) is higher than that of the expert (85.5%), and two out of five candidates detected box are cancerous, achieving a high precision while maintaining a high processing speed of 0.51 seconds/image. The proposed system showed 5.9 points higher on the F1 score compared to methods using the same object detection model and state-of-the-art data augmentation. In short, the system quickly and efficiently shows the radiologist where to look, greatly reducing the radiologist's workload.
Authors: Ziqing Ji, Pranav Kulkarni, Marko Neskovic, Kevin Nolan, Yan Xu
Abstract: With news and information being as easy to access as they currently are, it is more important than ever to ensure that people are not mislead by what they read. Recently, the rise of neural fake news (AI-generated fake news) and its demonstrated effectiveness at fooling humans has prompted the development of models to detect it. One such model is the Grover model, which can both detect neural fake news to prevent it, and generate it to demonstrate how a model could be misused to fool human readers. In this work we explore the Grover model's fake news detection capabilities by performing targeted attacks through perturbations on input news articles. Through this we test Grover's resilience to these adversarial attacks and expose some potential vulnerabilities which should be addressed in further iterations to ensure it can detect all types of fake news accurately.
Authors: Amr Hagag, Ahmed Gomaa, Dominik Kornek, Andreas Maier, Rainer Fietkau, Christoph Bert, Florian Putz, Yixing Huang
Abstract: Survival prediction for cancer patients is critical for optimal treatment selection and patient management. Current patient survival prediction methods typically extract survival information from patients' clinical record data or biological and imaging data. In practice, experienced clinicians can have a preliminary assessment of patients' health status based on patients' observable physical appearances, which are mainly facial features. However, such assessment is highly subjective. In this work, the efficacy of objectively capturing and using prognostic information contained in conventional portrait photographs using deep learning for survival predication purposes is investigated for the first time. A pre-trained StyleGAN2 model is fine-tuned on a custom dataset of our cancer patients' photos to empower its generator with generative ability suitable for patients' photos. The StyleGAN2 is then used to embed the photographs to its highly expressive latent space. Utilizing the state-of-the-art survival analysis models and based on StyleGAN's latent space photo embeddings, this approach achieved a C-index of 0.677, which is notably higher than chance and evidencing the prognostic value embedded in simple 2D facial images. In addition, thanks to StyleGAN's interpretable latent space, our survival prediction model can be validated for relying on essential facial features, eliminating any biases from extraneous information like clothing or background. Moreover, a health attribute is obtained from regression coefficients, which has important potential value for patient care.
Authors: Grzegorz Rype\'s\'c, Daniel Marczak, Sebastian Cygert, Tomasz Trzci\'nski, Bart{\l}omiej Twardowski
Abstract: Generalized Continual Category Discovery (GCCD) tackles learning from sequentially arriving, partially labeled datasets while uncovering new categories. Traditional methods depend on feature distillation to prevent forgetting the old knowledge. However, this strategy restricts the model's ability to adapt and effectively distinguish new categories. To address this, we introduce a novel technique integrating a learnable projector with feature distillation, thus enhancing model adaptability without sacrificing past knowledge. The resulting distribution shift of the previously learned categories is mitigated with the auxiliary category adaptation network. We demonstrate that while each component offers modest benefits individually, their combination - dubbed CAMP (Category Adaptation Meets Projected distillation) - significantly improves the balance between learning new information and retaining old. CAMP exhibits superior performance across several GCCD and Class Incremental Learning scenarios. The code is available at https://github.com/grypesc/CAMP.
Authors: Tsung-Wei Ke, Nikolaos Gkanatsios, Katerina Fragkiadaki
Abstract: Diffusion policies are conditional diffusion models that learn robot action distributions conditioned on the robot and environment state. They have recently shown to outperform both deterministic and alternative action distribution learning formulations. 3D robot policies use 3D scene feature representations aggregated from a single or multiple camera views using sensed depth. They have shown to generalize better than their 2D counterparts across camera viewpoints. We unify these two lines of work and present 3D Diffuser Actor, a neural policy equipped with a novel 3D denoising transformer that fuses information from the 3D visual scene, a language instruction and proprioception to predict the noise in noised 3D robot pose trajectories. 3D Diffuser Actor sets a new state-of-the-art on RLBench with an absolute performance gain of 18.1% over the current SOTA on a multi-view setup and an absolute gain of 13.1% on a single-view setup. On the CALVIN benchmark, it improves over the current SOTA by a 9% relative increase. It also learns to control a robot manipulator in the real world from a handful of demonstrations. Through thorough comparisons with the current SOTA policies and ablations of our model, we show 3D Diffuser Actor's design choices dramatically outperform 2D representations, regression and classification objectives, absolute attentions, and holistic non-tokenized 3D scene embeddings.
Authors: Abhijeet Parida, Daniel Capellan-Martin, Sara Atito, Muhammad Awais, Maria J. Ledesma-Carbayo, Marius G. Linguraru, Syed Muhammad Anwar
Abstract: Chest X-Ray (CXR) is a widely used clinical imaging modality and has a pivotal role in the diagnosis and prognosis of various lung and heart related conditions. Conventional automated clinical diagnostic tool design strategies relying on radiology reads and supervised learning, entail the cumbersome requirement of high quality annotated training data. To address this challenge, self-supervised pre-training has proven to outperform supervised pre-training in numerous downstream vision tasks, representing a significant breakthrough in the field. However, medical imaging pre-training significantly differs from pre-training with natural images (e.g., ImageNet) due to unique attributes of clinical images. In this context, we introduce Diverse Concept Modeling (DiCoM), a novel self-supervised training paradigm that leverages a student teacher framework for learning diverse concepts and hence effective representation of the CXR data. Hence, expanding beyond merely modeling a single primary label within an image, instead, effectively harnessing the information from all the concepts inherent in the CXR. The pre-trained model is subsequently fine-tuned to address diverse domain-specific tasks. Our proposed paradigm consistently demonstrates robust performance across multiple downstream tasks on multiple datasets, highlighting the success and generalizability of the pre-training strategy. To establish the efficacy of our methods we analyze both the power of learned representations and the speed of convergence (SoC) of our models. For diverse data and tasks, DiCoM is able to achieve in most cases better results compared to other state-of-the-art pre-training strategies. This when combined with the higher SoC and generalization capabilities positions DiCoM to be established as a foundation model for CXRs, a widely used imaging modality.
Authors: P. Bilha Githinji, Xi Yuan, Zhenglin Chen, Ijaz Gul, Dingqi Shang, Wen Liang, Jianming Deng, Dan Zeng, Dongmei yu, Chenggang Yan, Peiwu Qin
Abstract: Realizing sufficient separability between the distributions of healthy and pathological samples is a critical obstacle for pathology detection convolutional models. Moreover, these models exhibit a bias for contrast-based images, with diminished performance on texture-based medical images. This study introduces the notion of a population-level context for pathology detection and employs a graph theoretic approach to model and incorporate it into the latent code of an autoencoder via a refinement module we term PopuSense. PopuSense seeks to capture additional intra-group variations inherent in biomedical data that a local or global context of the convolutional model might miss or smooth out. Proof-of-concept experiments on contrast-based and texture-based images, with minimal adaptation, encounter the existing preference for intensity-based input. Nevertheless, PopuSense demonstrates improved separability in contrast-based images, presenting an additional avenue for refining representations learned by a model.
Authors: Jonas Schramm, Niclas V\"odisch, K\"ursat Petek, B Ravi Kiran, Senthil Yogamani, Wolfram Burgard, Abhinav Valada
Abstract: Semantic scene segmentation from a bird's-eye-view (BEV) perspective plays a crucial role in facilitating planning and decision-making for mobile robots. Although recent vision-only methods have demonstrated notable advancements in performance, they often struggle under adverse illumination conditions such as rain or nighttime. While active sensors offer a solution to this challenge, the prohibitively high cost of LiDARs remains a limiting factor. Fusing camera data with automotive radars poses a more inexpensive alternative but has received less attention in prior research. In this work, we aim to advance this promising avenue by introducing BEVCar, a novel approach for joint BEV object and map segmentation. The core novelty of our approach lies in first learning a point-based encoding of raw radar data, which is then leveraged to efficiently initialize the lifting of image features into the BEV space. We perform extensive experiments on the nuScenes dataset and demonstrate that BEVCar outperforms the current state of the art. Moreover, we show that incorporating radar information significantly enhances robustness in challenging environmental conditions and improves segmentation performance for distant objects. To foster future research, we provide the weather split of the nuScenes dataset used in our experiments, along with our code and trained models at http://bevcar.cs.uni-freiburg.de.
Authors: Minyoung Huh, Brian Cheung, Tongzhou Wang, Phillip Isola
Abstract: We argue that representations in AI models, particularly deep networks, are converging. First, we survey many examples of convergence in the literature: over time and across multiple domains, the ways by which different neural networks represent data are becoming more aligned. Next, we demonstrate convergence across data modalities: as vision models and language models get larger, they measure distance between datapoints in a more and more alike way. We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality. We term such a representation the platonic representation and discuss several possible selective pressures toward it. Finally, we discuss the implications of these trends, their limitations, and counterexamples to our analysis.
Authors: Yixing Huang, Zahra Khodabakhshi, Ahmed Gomaa, Manuel Schmidt, Rainer Fietkau, Matthias Guckenberger, Nicolaus Andratschke, Christoph Bert, Stephanie Tanadini-Lang, Florian Putz
Abstract: Objectives: This work aims to explore the impact of multicenter data heterogeneity on deep learning brain metastases (BM) autosegmentation performance, and assess the efficacy of an incremental transfer learning technique, namely learning without forgetting (LWF), to improve model generalizability without sharing raw data. Materials and methods: A total of six BM datasets from University Hospital Erlangen (UKER), University Hospital Zurich (USZ), Stanford, UCSF, NYU and BraTS Challenge 2023 on BM segmentation were used for this evaluation. First, the multicenter performance of a convolutional neural network (DeepMedic) for BM autosegmentation was established for exclusive single-center training and for training on pooled data, respectively. Subsequently bilateral collaboration was evaluated, where a UKER pretrained model is shared to another center for further training using transfer learning (TL) either with or without LWF. Results: For single-center training, average F1 scores of BM detection range from 0.625 (NYU) to 0.876 (UKER) on respective single-center test data. Mixed multicenter training notably improves F1 scores at Stanford and NYU, with negligible improvement at other centers. When the UKER pretrained model is applied to USZ, LWF achieves a higher average F1 score (0.839) than naive TL (0.570) and single-center training (0.688) on combined UKER and USZ test data. Naive TL improves sensitivity and contouring accuracy, but compromises precision. Conversely, LWF demonstrates commendable sensitivity, precision and contouring accuracy. When applied to Stanford, similar performance was observed. Conclusion: Data heterogeneity results in varying performance in BM autosegmentation, posing challenges to model generalizability. LWF is a promising approach to peer-to-peer privacy-preserving model training.
Authors: Chen Wang, Kaiyi Ji, Junyi Geng, Zhongqiang Ren, Taimeng Fu, Fan Yang, Yifan Guo, Haonan He, Xiangyu Chen, Zitong Zhan, Qiwei Du, Shaoshu Su, Bowen Li, Yuheng Qiu, Yi Du, Qihang Li, Yifan Yang, Xiao Lin, Zhipeng Zhao
Abstract: Data-driven methods such as reinforcement and imitation learning have achieved remarkable success in robot autonomy. However, their data-centric nature still hinders them from generalizing well to ever-changing environments. Moreover, collecting large datasets for robotic tasks is often impractical and expensive. To overcome these challenges, we introduce a new self-supervised neural-symbolic (NeSy) computational framework, imperative learning (IL), for robot autonomy, leveraging the generalization abilities of symbolic reasoning. The framework of IL consists of three primary components: a neural module, a reasoning engine, and a memory system. We formulate IL as a special bilevel optimization (BLO), which enables reciprocal learning over the three modules. This overcomes the label-intensive obstacles associated with data-driven approaches and takes advantage of symbolic reasoning concerning logical reasoning, physical principles, geometric analysis, etc. We discuss several optimization techniques for IL and verify their effectiveness in five distinct robot autonomy tasks including path planning, rule induction, optimal control, visual odometry, and multi-robot routing. Through various experiments, we show that IL can significantly enhance robot autonomy capabilities and we anticipate that it will catalyze further research across diverse domains.
Authors: Haibo Jin, Leyang Hu, Xinuo Li, Peiyan Zhang, Chonghan Chen, Jun Zhuang, Haohan Wang
Abstract: The rapid evolution of artificial intelligence (AI) through developments in Large Language Models (LLMs) and Vision-Language Models (VLMs) has brought significant advancements across various technological domains. While these models enhance capabilities in natural language processing and visual interactive tasks, their growing adoption raises critical concerns regarding security and ethical alignment. This survey provides an extensive review of the emerging field of jailbreaking--deliberately circumventing the ethical and operational boundaries of LLMs and VLMs--and the consequent development of defense mechanisms. Our study categorizes jailbreaks into seven distinct types and elaborates on defense strategies that address these vulnerabilities. Through this comprehensive examination, we identify research gaps and propose directions for future studies to enhance the security frameworks of LLMs and VLMs. Our findings underscore the necessity for a unified perspective that integrates both jailbreak strategies and defensive solutions to foster a robust, secure, and reliable environment for the next generation of language models. More details can be found on our website: \url{https://chonghan-chen.com/llm-jailbreak-zoo-survey/}.
Authors: Chaoxing Huang, Ziqiang Yu, Zijian Gao, Qiuyi Shen, Queenie Chan, Vincent Wai-Sun Wong, Winnie Chiu-Wing Chu, Weitian Chen
Abstract: This study evaluated a deep learning-based method using Deep Image Prior (DIP) to quantify triglyceride double bonds from chemical-shift encoded multi-echo gradient echo images without network training. We employed a cost function based on signal constraints to iteratively update the neural network on a single dataset. The method was validated using phantom experiments and in vivo scans. Results showed close alignment between measured and reference double bond values, with phantom experiments yielding a Pearson correlation coefficient of 0.96 (p = .0005). In vivo results demonstrated good agreement in subcutaneous fat. We conclude that Deep Image Prior shows feasibility for quantifying double bonds and fatty acid content from chemical-shift encoded multi-echo MRI.
Authors: Pooyan Rahmanzadehgervi, Logan Bolton, Mohammad Reza Taesiri, Anh Totti Nguyen
Abstract: While large language models with vision capabilities (VLMs), e.g., GPT-4o and Gemini 1.5 Pro, are powering various image-text applications and scoring high on many vision-understanding benchmarks, we find that they are surprisingly still struggling with low-level vision tasks that are easy to humans. Specifically, on BlindTest, our suite of 7 very simple tasks such as identifying (a) whether two circles overlap; (b) whether two lines intersect; (c) which letter is being circled in a word; and (d) counting circles in an Olympic-like logo, four state-of-the-art VLMs are only 58.57% accurate on average. Claude 3.5 Sonnet performs the best at 74.01% accuracy, but this is still far from the human expected accuracy of 100%. Across different image resolutions and line widths, VLMs consistently struggle with tasks that require precise spatial information and recognizing geometric primitives that overlap or are close together. Code and data are available at: https://vlmsareblind.github.io
Authors: Wei Dai, Rui Liu, Zixuan Wu, Tianyi Wu, Min Wang, Junxian Zhou, Yixuan Yuan, Jun Liu
Abstract: Early detection and accurate diagnosis can predict the risk of malignant disease transformation, thereby increasing the probability of effective treatment. A mild syndrome with small infected regions is an ominous warning and is foremost in the early diagnosis of diseases. Deep learning algorithms, such as convolutional neural networks (CNNs), have been used to segment natural or medical objects, showing promising results. However, analyzing medical objects of small areas in images remains a challenge due to information losses and compression defects caused by convolution and pooling operations in CNNs. These losses and defects become increasingly significant as the network deepens, particularly for small medical objects. To address these challenges, we propose a novel scale-variant attention-based network (SvANet) for accurate small-scale object segmentation in medical images. The SvANet consists of Monte Carlo attention, scale-variant attention, and vision transformer, which incorporates cross-scale features and alleviates compression artifacts for enhancing the discrimination of small medical objects. Quantitative experimental results demonstrate the superior performance of SvANet, achieving 96.12%, 96.11%, 89.79%, 84.15%, 80.25%, 73.05%, and 72.58% in mean Dice coefficient for segmenting kidney tumors, skin lesions, hepatic tumors, polyps, surgical excision cells, retinal vasculatures, and sperms, which occupy less than 1% of the image areas in KiTS23, ISIC 2018, ATLAS, PolypGen, TissueNet, FIVES, and SpermHealth datasets, respectively.
Authors: Toan Nguyen, Minh Nhat Vu, Baoru Huang, An Vuong, Quan Vuong, Ngan Le, Thieu Vo, Anh Nguyen
Abstract: 6-DoF grasp detection has been a fundamental and challenging problem in robotic vision. While previous works have focused on ensuring grasp stability, they often do not consider human intention conveyed through natural language, hindering effective collaboration between robots and users in complex 3D environments. In this paper, we present a new approach for language-driven 6-DoF grasp detection in cluttered point clouds. We first introduce Grasp-Anything-6D, a large-scale dataset for the language-driven 6-DoF grasp detection task with 1M point cloud scenes and more than 200M language-associated 3D grasp poses. We further introduce a novel diffusion model that incorporates a new negative prompt guidance learning strategy. The proposed negative prompt strategy directs the detection process toward the desired object while steering away from unwanted ones given the language input. Our method enables an end-to-end framework where humans can command the robot to grasp desired objects in a cluttered scene using natural language. Intensive experimental results show the effectiveness of our method in both benchmarking experiments and real-world scenarios, surpassing other baselines. In addition, we demonstrate the practicality of our approach in real-world robotic applications. Our project is available at https://airvlab.github.io/grasp-anything.
Authors: Nikolaos Ntampakis, Konstantinos Diamantaras, Ioanna Chouvarda, Vasileios Argyriou, Panagiotis Sarigianndis
Abstract: Dementia, a debilitating neurological condition affecting millions worldwide, presents significant diagnostic challenges. In this work, we introduce a novel methodology for the classification of demented and non-demented elderly patients using 3D brain Magnetic Resonance Imaging (MRI) scans. Our approach features a unique technique for selectively processing MRI slices, focusing on the most relevant brain regions and excluding less informative sections. This methodology is complemented by a confidence-based classification committee composed of three custom deep learning models: Dem3D ResNet, Dem3D CNN, and Dem3D EfficientNet. These models work synergistically to enhance decision-making accuracy, leveraging their collective strengths. Tested on the Open Access Series of Imaging Studies(OASIS) dataset, our method achieved an impressive accuracy of 94.12%, surpassing existing methodologies. Furthermore, validation on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset confirmed the robustness and generalizability of our approach. The use of explainable AI (XAI) techniques and comprehensive ablation studies further substantiate the effectiveness of our techniques, providing insights into the decision-making process and the importance of our methodology. This research offers a significant advancement in dementia diagnosis, providing a highly accurate and efficient tool for clinical applications.
Authors: Vito Paolo Pastore, Massimiliano Ciranni, Davide Marinelli, Francesca Odone, Vittorio Murino
Abstract: It is widely recognized that deep neural networks are sensitive to bias in the data. This means that during training these models are likely to learn spurious correlations between data and labels, resulting in limited generalization abilities and low performance. In this context, model debiasing approaches can be devised aiming at reducing the model's dependency on such unwanted correlations, either leveraging the knowledge of bias information or not. In this work, we focus on the latter and more realistic scenario, showing the importance of accurately predicting the bias-conflicting and bias-aligned samples to obtain compelling performance in bias mitigation. On this ground, we propose to conceive the problem of model bias from an out-of-distribution perspective, introducing a new bias identification method based on anomaly detection. We claim that when data is mostly biased, bias-conflicting samples can be regarded as outliers with respect to the bias-aligned distribution in the feature space of a biased model, thus allowing for precisely detecting them with an anomaly detection method. Coupling the proposed bias identification approach with bias-conflicting data upsampling and augmentation in a two-step strategy, we reach state-of-the-art performance on synthetic and real benchmark datasets. Ultimately, our proposed approach shows that the data bias issue does not necessarily require complex debiasing methods, given that an accurate bias identification procedure is defined.