Authors: Hua Yan, Heng Tan, Yi Ding, Peifei Zhou, Vinod Namboodiri, Yu Yang
Abstract: Human Activity Recognition (HAR) using Inertial Measurement Unit (IMU) sensors is critical for applications in healthcare, safety, and industrial production. However, variations in activity patterns, device types, and sensor placements create distribution gaps across datasets, reducing the performance of HAR models. To address this, we propose LanHAR, a novel system that leverages Large Language Models (LLMs) to generate semantic interpretations of sensor readings and activity labels for cross-dataset HAR. This approach not only mitigates cross-dataset heterogeneity but also enhances the recognition of new activities. LanHAR employs an iterative re-generation method to produce high-quality semantic interpretations with LLMs and a two-stage training framework that bridges the semantic interpretations of sensor readings and activity labels. This ultimately leads to a lightweight sensor encoder suitable for mobile deployment, enabling any sensor reading to be mapped into the semantic interpretation space. Experiments on four public datasets demonstrate that our approach significantly outperforms state-of-the-art methods in both cross-dataset HAR and new activity recognition. The source code will be made publicly available.
Authors: Alejandro Aparcedo, Christian Lopez, Abhinav Kotta, Mengjie Li
Abstract: Extreme weather events are increasingly common due to climate change, posing significant risks. To mitigate further damage, a shift towards renewable energy is imperative. Unfortunately, underrepresented communities that are most affected often receive infrastructure improvements last. We propose a novel visual spatiotemporal framework for predicting nighttime lights (NTL), power outage severity and location before and after major hurricanes. Central to our solution is the Visual-Spatiotemporal Graph Neural Network (VST-GNN), to learn spatial and temporal coherence from images. Our work brings awareness to underrepresented areas in urgent need of enhanced energy infrastructure, such as future photovoltaic (PV) deployment. By identifying the severity and localization of power outages, our initiative aims to raise awareness and prompt action from policymakers and community stakeholders. Ultimately, this effort seeks to empower regions with vulnerable energy infrastructure, enhancing resilience and reliability for at-risk communities.
Authors: Zhen Han, Zeyinzi Jiang, Yulin Pan, Jingfeng Zhang, Chaojie Mao, Chenwei Xie, Yu Liu, Jingren Zhou
Abstract: Diffusion models have emerged as a powerful generative technology and have been found to be applicable in various scenarios. Most existing foundational diffusion models are primarily designed for text-guided visual generation and do not support multi-modal conditions, which are essential for many visual editing tasks. This limitation prevents these foundational diffusion models from serving as a unified model in the field of visual generation, like GPT-4 in the natural language processing field. In this work, we propose ACE, an All-round Creator and Editor, which achieves comparable performance compared to those expert models in a wide range of visual generation tasks. To achieve this goal, we first introduce a unified condition format termed Long-context Condition Unit (LCU), and propose a novel Transformer-based diffusion model that uses LCU as input, aiming for joint training across various generation and editing tasks. Furthermore, we propose an efficient data collection approach to address the issue of the absence of available training data. It involves acquiring pairwise images with synthesis-based or clustering-based pipelines and supplying these pairs with accurate textual instructions by leveraging a fine-tuned multi-modal large language model. To comprehensively evaluate the performance of our model, we establish a benchmark of manually annotated pairs data across a variety of visual generation tasks. The extensive experimental results demonstrate the superiority of our model in visual generation fields. Thanks to the all-in-one capabilities of our model, we can easily build a multi-modal chat system that responds to any interactive request for image creation using a single model to serve as the backend, avoiding the cumbersome pipeline typically employed in visual agents. Code and models will be available on the project page: https://ali-vilab.github.io/ace-page/.
Authors: Jiachen Ye, Dingyu Wang, Shaocheng Jia, Xin Pei, Zi Yang, Yi Zhang, S. C. Wong
Abstract: Real-time estimation of vehicle locations and speeds is crucial for developing many beneficial transportation applications in traffic management and control, e.g., adaptive signal control. Recent advances in communication technologies facilitate the emergence of connected vehicles (CVs), which can share traffic information with nearby CVs or infrastructures. At the early stage of connectivity, only a portion of vehicles are CVs. The locations and speeds for those non-CVs (NCs) are not accessible and must be estimated to obtain the full traffic information. To address the above problem, this paper proposes a novel CV-based Vehicle Location and Speed estimation network, CVVLSNet, to simultaneously estimate the vehicle locations and speeds exclusively using partial CV trajectory data. A road cell occupancy (RCO) method is first proposed to represent the variable vehicle state information. Spatiotemporal interactions can be integrated by simply fusing the RCO representations. Then, CVVLSNet, taking the Coding-RAte TransformEr (CRATE) network as a backbone, is introduced to estimate the vehicle locations and speeds. Moreover, physical vehicle size constraints are also considered in loss functions. Extensive experiments indicate that the proposed method significantly outperformed the existing method under various CV penetration rates, signal timings, and volume-to-capacity ratios.
Authors: Hongyu Chen, Weiming Zeng, Chengcheng Chen, Luhui Cai, Fei Wang, Lei Wang, Wei Zhang, Yueyang Li, Hongjie Yan, Wai Ting Siok, Nizhuan Wang
Abstract: In the fields of affective computing (AC) and brain-machine interface (BMI), the analysis of physiological and behavioral signals to discern individual emotional states has emerged as a critical research frontier. While deep learning-based approaches have made notable strides in EEG emotion recognition, particularly in feature extraction and pattern recognition, significant challenges persist in achieving end-to-end emotion computation, including real-time processing, individual adaptation, and seamless user interaction. This paper presents the EEG Emotion Copilot, a system leveraging a lightweight large language model (LLM) operating in a local setting. The system is designed to first recognize emotional states directly from EEG signals, subsequently generate personalized diagnostic and treatment suggestions, and finally support the automation of electronic medical records. The proposed solution emphasizes both the accuracy of emotion recognition and an enhanced user experience, facilitated by an intuitive interface for participant interaction. We further discuss the construction of the data framework, model pruning, training, and deployment strategies aimed at improving real-time performance and computational efficiency. Privacy concerns are also addressed, with a focus on ethical data collection, processing, and the protection of users' personal information. Through these efforts, we aim to advance the application of AC in the medical domain, offering innovative approaches to mental health diagnostics and treatment.
Authors: Yi-Hao Peng, Faria Huq, Yue Jiang, Jason Wu, Amanda Xin Yue Li, Jeffrey Bigham, Amy Pavel
Abstract: Enabling machines to understand structured visuals like slides and user interfaces is essential for making them accessible to people with disabilities. However, achieving such understanding computationally has required manual data collection and annotation, which is time-consuming and labor-intensive. To overcome this challenge, we present a method to generate synthetic, structured visuals with target labels using code generation. Our method allows people to create datasets with built-in labels and train models with a small number of human-annotated examples. We demonstrate performance improvements in three tasks for understanding slides and UIs: recognizing visual elements, describing visual content, and classifying visual content types.
Authors: Saihui Hou, Panjian Huang, Zengbin Wang, Yuan Liu, Zeyu Li, Man Zhang, Yongzhen Huang
Abstract: This paper addresses the challenge of animal re-identification, an emerging field that shares similarities with person re-identification but presents unique complexities due to the diverse species, environments and poses. To facilitate research in this domain, we introduce OpenAnimals, a flexible and extensible codebase designed specifically for animal re-identification. We conduct a comprehensive study by revisiting several state-of-the-art person re-identification methods, including BoT, AGW, SBS, and MGN, and evaluate their effectiveness on animal re-identification benchmarks such as HyenaID, LeopardID, SeaTurtleID, and WhaleSharkID. Our findings reveal that while some techniques generalize well, many do not, underscoring the significant differences between the two tasks. To bridge this gap, we propose ARBase, a strong \textbf{Base} model tailored for \textbf{A}nimal \textbf{R}e-identification, which incorporates insights from extensive experiments and introduces simple yet effective animal-oriented designs. Experiments demonstrate that ARBase consistently outperforms existing baselines, achieving state-of-the-art performance across various benchmarks.
Authors: Anna Deichler, Jim O'Regan, Jonas Beskow
Abstract: In this paper, we present a novel dataset captured using a VR headset to record conversations between participants within a physics simulator (AI2-THOR). Our primary objective is to extend the field of co-speech gesture generation by incorporating rich contextual information within referential settings. Participants engaged in various conversational scenarios, all based on referential communication tasks. The dataset provides a rich set of multimodal recordings such as motion capture, speech, gaze, and scene graphs. This comprehensive dataset aims to enhance the understanding and development of gesture generation models in 3D scenes by providing diverse and contextually rich data.
Authors: Jian Shi, Zhenyu Li, Peter Wonka
Abstract: We introduce \textit{ImmersePro}, an innovative framework specifically designed to transform single-view videos into stereo videos. This framework utilizes a novel dual-branch architecture comprising a disparity branch and a context branch on video data by leveraging spatial-temporal attention mechanisms. \textit{ImmersePro} employs implicit disparity guidance, enabling the generation of stereo pairs from video sequences without the need for explicit disparity maps, thus reducing potential errors associated with disparity estimation models. In addition to the technical advancements, we introduce the YouTube-SBS dataset, a comprehensive collection of 423 stereo videos sourced from YouTube. This dataset is unprecedented in its scale, featuring over 7 million stereo pairs, and is designed to facilitate training and benchmarking of stereo video generation models. Our experiments demonstrate the effectiveness of \textit{ImmersePro} in producing high-quality stereo videos, offering significant improvements over existing methods. Compared to the best competitor stereo-from-mono we quantitatively improve the results by 11.76\% (L1), 6.39\% (SSIM), and 5.10\% (PSNR).
Authors: Kun Yuan, Vinkle Srivastav, Nassir Navab, Nicolas Padoy
Abstract: Surgical video-language pretraining (VLP) faces unique challenges due to the knowledge domain gap and the scarcity of multi-modal data. This study aims to bridge the gap by addressing issues regarding textual information loss in surgical lecture videos and the spatial-temporal challenges of surgical VLP. We propose a hierarchical knowledge augmentation approach and a novel Procedure-Encoded Surgical Knowledge-Augmented Video-Language Pretraining (PeskaVLP) framework to tackle these issues. The knowledge augmentation uses large language models (LLM) for refining and enriching surgical concepts, thus providing comprehensive language supervision and reducing the risk of overfitting. PeskaVLP combines language supervision with visual self-supervision, constructing hard negative samples and employing a Dynamic Time Warping (DTW) based loss function to effectively comprehend the cross-modal procedural alignment. Extensive experiments on multiple public surgical scene understanding and cross-modal retrieval datasets show that our proposed method significantly improves zero-shot transferring performance and offers a generalist visual representation for further advancements in surgical scene understanding.
Authors: Aleyna K\"ut\"uk, Tevfik Metin Sezgin
Abstract: Scene sketch semantic segmentation is a crucial task for various applications including sketch-to-image retrieval and scene understanding. Existing sketch segmentation methods treat sketches as bitmap images, leading to the loss of temporal order among strokes due to the shift from vector to image format. Moreover, these methods struggle to segment objects from categories absent in the training data. In this paper, we propose a Class-Agnostic Visio-Temporal Network (CAVT) for scene sketch semantic segmentation. CAVT employs a class-agnostic object detector to detect individual objects in a scene and groups the strokes of instances through its post-processing module. This is the first approach that performs segmentation at both the instance and stroke levels within scene sketches. Furthermore, there is a lack of free-hand scene sketch datasets with both instance and stroke-level class annotations. To fill this gap, we collected the largest Free-hand Instance- and Stroke-level Scene Sketch Dataset (FrISS) that contains 1K scene sketches and covers 403 object classes with dense annotations. Extensive experiments on FrISS and other datasets demonstrate the superior performance of our method over state-of-the-art scene sketch segmentation models. The code and dataset will be made public after acceptance.
Authors: Sachin Karmani, Thanushon Sivakaran, Gaurav Prasad, Mehmet Ali, Wenbo Yang, Sheyang Tang
Abstract: Deep learning models often function as black boxes, providing no straightforward reasoning for their predictions. This is particularly true for computer vision models, which process tensors of pixel values to generate outcomes in tasks such as image classification and object detection. To elucidate the reasoning of these models, class activation maps (CAMs) are used to highlight salient regions that influence a model's output. This research introduces KPCA-CAM, a technique designed to enhance the interpretability of Convolutional Neural Networks (CNNs) through improved class activation maps. KPCA-CAM leverages Principal Component Analysis (PCA) with the kernel trick to capture nonlinear relationships within CNN activations more effectively. By mapping data into higher-dimensional spaces with kernel functions and extracting principal components from this transformed hyperplane, KPCA-CAM provides more accurate representations of the underlying data manifold. This enables a deeper understanding of the features influencing CNN decisions. Empirical evaluations on the ILSVRC dataset across different CNN models demonstrate that KPCA-CAM produces more precise activation maps, providing clearer insights into the model's reasoning compared to existing CAM algorithms. This research advances CAM techniques, equipping researchers and practitioners with a powerful tool to gain deeper insights into CNN decision-making processes and overall behaviors.
Authors: Rohaifa Khaldi, Domingo Alcaraz-Segura, Ignacio S\'anchez-Herrera, Javier Martinez-Lopez, Carlos Javier Navarro, Siham Tabik
Abstract: Social media images have shown to be a valuable source of information for understanding human interactions with important subjects such as cultural heritage, biodiversity and nature among others. The task of grouping such images into a number of semantically meaningful clusters without labels is challenging given the high diversity and complex nature of the visual content of these images in addition to their large volume. On the other hand, the last advances in Large Visual Models (LVM), Large Language Models (LLM) and Large Visual Language Models (LVLM) provide an important opportunity to explore new productive and scalable solutions. This works proposes, analyzes, and compares various approaches based on one or more state-of-the art LVM, LLM and LVLM, for mapping social media images into a number of pre-defined classes. As case study, we consider the problem of understanding the interactions between human and nature, also known as Nature's Contribution to People or Cultural Ecosystem Services (CES). Our experiments reveal that the top-performing approaches, delivering highly competitive results, are the fine-tuned LVM DINOv2 on a small labeled dataset and LVLM models like the proprietary GPT-4 (gpt-4o-mini) using a simple prompt.
Authors: Mohssen E. Elshaar, Zeyad M. Manaa, Mohammed R. Elbalshy, Abdul Jabbar Siddiqui, Ayman M. Abdallah
Abstract: Unmanned Aerial Vehicles (UAVs) are becoming more popular in various sectors, offering many benefits, yet introducing significant challenges to privacy and safety. This paper investigates state-of-the-art solutions for detecting and tracking quadrotor UAVs to address these concerns. Cutting-edge deep learning models, specifically the YOLOv5 and YOLOv8 series, are evaluated for their performance in identifying UAVs accurately and quickly. Additionally, robust tracking systems, BoT-SORT and Byte Track, are integrated to ensure reliable monitoring even under challenging conditions. Our tests on the DUT dataset reveal that while YOLOv5 models generally outperform YOLOv8 in detection accuracy, the YOLOv8 models excel in recognizing less distinct objects, demonstrating their adaptability and advanced capabilities. Furthermore, BoT-SORT demonstrated superior performance over Byte Track, achieving higher IoU and lower center error in most cases, indicating more accurate and stable tracking. Code: https://github.com/zmanaa/UAV_detection_and_tracking Tracking demo: https://drive.google.com/file/d/1pe6HC5kQrgTbA2QrjvMN-yjaZyWeAvDT/view?usp=sharing
URLs: https://github.com/zmanaa/UAV_detection_and_tracking, https://drive.google.com/file/d/1pe6HC5kQrgTbA2QrjvMN-yjaZyWeAvDT/view?usp=sharing
Authors: Dasong Li, Wenjie Li, Baili Lu, Hongsheng Li, Sizhuo Ma, Gurunandan Krishnan, Jian Wang
Abstract: Understanding and modeling the popularity of User Generated Content (UGC) short videos on social media platforms presents a critical challenge with broad implications for content creators and recommendation systems. This study delves deep into the intricacies of predicting engagement for newly published videos with limited user interactions. Surprisingly, our findings reveal that Mean Opinion Scores from previous video quality assessment datasets do not strongly correlate with video engagement levels. To address this, we introduce a substantial dataset comprising 90,000 real-world UGC short videos from Snapchat. Rather than relying on view count, average watch time, or rate of likes, we propose two metrics: normalized average watch percentage (NAWP) and engagement continuation rate (ECR) to describe the engagement levels of short videos. Comprehensive multi-modal features, including visual content, background music, and text data, are investigated to enhance engagement prediction. With the proposed dataset and two key metrics, our method demonstrates its ability to predict engagements of short videos purely from video content.
Authors: Zhangshuo Qi, Junyi Ma, Jingyi Xu, Zijie Zhou, Luqi Cheng, Guangming Xiong
Abstract: Place recognition is a crucial module to ensure autonomous vehicles obtain usable localization information in GPS-denied environments. In recent years, multimodal place recognition methods have gained increasing attention due to their ability to overcome the weaknesses of unimodal sensor systems by leveraging complementary information from different modalities. However, challenges arise from the necessity of harmonizing data across modalities and exploiting the spatio-temporal correlations between them sufficiently. In this paper, we propose a 3D Gaussian Splatting-based multimodal place recognition neural network dubbed GSPR. It explicitly combines multi-view RGB images and LiDAR point clouds into a spatio-temporally unified scene representation with the proposed Multimodal Gaussian Splatting. A network composed of 3D graph convolution and transformer is designed to extract high-level spatio-temporal features and global descriptors from the Gaussian scenes for place recognition. We evaluate our method on the nuScenes dataset, and the experimental results demonstrate that our method can effectively leverage complementary strengths of both multi-view cameras and LiDAR, achieving SOTA place recognition performance while maintaining solid generalization ability. Our open-source code is available at https://github.com/QiZS-BIT/GSPR.
Authors: Moinak Bhattacharya, Gagandeep Singh, Shubham Jain, Prateek Prasanna
Abstract: In this work, we present RadGazeGen, a novel framework for integrating experts' eye gaze patterns and radiomic feature maps as controls to text-to-image diffusion models for high fidelity medical image generation. Despite the recent success of text-to-image diffusion models, text descriptions are often found to be inadequate and fail to convey detailed disease-specific information to these models to generate clinically accurate images. The anatomy, disease texture patterns, and location of the disease are extremely important to generate realistic images; moreover the fidelity of image generation can have significant implications in downstream tasks involving disease diagnosis or treatment repose assessment. Hence, there is a growing need to carefully define the controls used in diffusion models for medical image generation. Eye gaze patterns of radiologists are important visuo-cognitive information, indicative of subtle disease patterns and spatial location. Radiomic features further provide important subvisual cues regarding disease phenotype. In this work, we propose to use these gaze patterns in combination with standard radiomics descriptors, as controls, to generate anatomically correct and disease-aware medical images. RadGazeGen is evaluated for image generation quality and diversity on the REFLACX dataset. To demonstrate clinical applicability, we also show classification performance on the generated images from the CheXpert test set (n=500) and long-tailed learning performance on the MIMIC-CXR-LT test set (n=23550).
Authors: Laura Bravo-S\'anchez, Jaewoo Heo, Zhenzhen Weng, Kuan-Chieh Wang, Serena Yeung-Levy
Abstract: Social dynamics in close human interactions pose significant challenges for Human Mesh Estimation (HME), particularly due to the complexity of physical contacts and the scarcity of training data. Addressing these challenges, we introduce a novel data generation method that utilizes Large Vision Language Models (LVLMs) to annotate contact maps which guide test-time optimization to produce paired image and pseudo-ground truth meshes. This methodology not only alleviates the annotation burden but also enables the assembly of a comprehensive dataset specifically tailored for close interactions in HME. Our Ask Pose Unite (APU) dataset, comprising over 6.2k human mesh pairs in contact covering diverse interaction types, is curated from images depicting naturalistic person-to-person scenes. We empirically show that using our dataset to train a diffusion-based contact prior, used as guidance during optimization, improves mesh estimation on unseen interactions. Our work addresses longstanding challenges of data scarcity for close interactions in HME enhancing the field's capabilities of handling complex interaction scenarios.
Authors: Qihang Zhou, Jiangtao Yan, Shibo He, Wenchao Meng, Jiming Chen
Abstract: Zero-shot (ZS) 3D anomaly detection is a crucial yet unexplored field that addresses scenarios where target 3D training samples are unavailable due to practical concerns like privacy protection. This paper introduces PointAD, a novel approach that transfers the strong generalization capabilities of CLIP for recognizing 3D anomalies on unseen objects. PointAD provides a unified framework to comprehend 3D anomalies from both points and pixels. In this framework, PointAD renders 3D anomalies into multiple 2D renderings and projects them back into 3D space. To capture the generic anomaly semantics into PointAD, we propose hybrid representation learning that optimizes the learnable text prompts from 3D and 2D through auxiliary point clouds. The collaboration optimization between point and pixel representations jointly facilitates our model to grasp underlying 3D anomaly patterns, contributing to detecting and segmenting anomalies of unseen diverse 3D objects. Through the alignment of 3D and 2D space, our model can directly integrate RGB information, further enhancing the understanding of 3D anomalies in a plug-and-play manner. Extensive experiments show the superiority of PointAD in ZS 3D anomaly detection across diverse unseen objects.
Authors: Chieh-Yun Chen, Li-Wu Tsao, Chiang Tseng, Hong-Han Shuai
Abstract: This paper analyzes the impact of causal manner in the text encoder of text-to-image (T2I) diffusion models, which can lead to information bias and loss. Previous works have focused on addressing the issues through the denoising process. However, there is no research discussing how text embedding contributes to T2I models, especially when generating more than one object. In this paper, we share a comprehensive analysis of text embedding: i) how text embedding contributes to the generated images and ii) why information gets lost and biases towards the first-mentioned object. Accordingly, we propose a simple but effective text embedding balance optimization method, which is training-free, with an improvement of 90.05% on information balance in stable diffusion. Furthermore, we propose a new automatic evaluation metric that quantifies information loss more accurately than existing methods, achieving 81% concordance with human assessments. This metric effectively measures the presence and accuracy of objects, addressing the limitations of current distribution scores like CLIP's text-image similarities.
Authors: Leheng Li, Weichao Qiu, Yingjie Cai, Xu Yan, Qing Lian, Bingbing Liu, Ying-Cong Chen
Abstract: The advancement of autonomous driving is increasingly reliant on high-quality annotated datasets, especially in the task of 3D occupancy prediction, where the occupancy labels require dense 3D annotation with significant human effort. In this paper, we propose SyntheOcc, which denotes a diffusion model that Synthesize photorealistic and geometric-controlled images by conditioning Occupancy labels in driving scenarios. This yields an unlimited amount of diverse, annotated, and controllable datasets for applications like training perception models and simulation. SyntheOcc addresses the critical challenge of how to efficiently encode 3D geometric information as conditional input to a 2D diffusion model. Our approach innovatively incorporates 3D semantic multi-plane images (MPIs) to provide comprehensive and spatially aligned 3D scene descriptions for conditioning. As a result, SyntheOcc can generate photorealistic multi-view images and videos that faithfully align with the given geometric labels (semantics in 3D voxel space). Extensive qualitative and quantitative evaluations of SyntheOcc on the nuScenes dataset prove its effectiveness in generating controllable occupancy datasets that serve as an effective data augmentation to perception models.
Authors: Huy Nguyen, Kien Nguyen, Akila Pemasiri, Sridha Sridharan, Clinton Fookes
Abstract: This study introduces a new framework for 3D person re-identification (re-ID) that leverages readily available high-resolution texture data in 3D reconstruction to improve the performance and explainability of the person re-ID task. We propose a method to emphasize texture in 3D person re-ID models by incorporating UVTexture mapping, which better differentiates human subjects. Our approach uniquely combines UVTexture and its heatmaps with 3D models to visualize and explain the person re-ID process. In particular, the visualization and explanation are achieved through activation maps and attribute-based attention maps, which highlight the important regions and features contributing to the person re-ID decision. Our contributions include: (1) a novel technique for emphasizing texture in 3D models using UVTexture processing, (2) an innovative method for explicating person re-ID matches through a combination of 3D models and UVTexture mapping, and (3) achieving state-of-the-art performance in 3D person re-ID. We ensure the reproducibility of our results by making all data, codes, and models publicly available.
Authors: Changlin Li, Jiawei Zhang, Sihao Lin, Zongxin Yang, Junwei Liang, Xiaodan Liang, Xiaojun Chang
Abstract: The rapid advancements in Large Vision Models (LVMs), such as Vision Transformers (ViTs) and diffusion models, have led to an increasing demand for computational resources, resulting in substantial financial and environmental costs. This growing challenge highlights the necessity of developing efficient training methods for LVMs. Progressive learning, a training strategy in which model capacity gradually increases during training, has shown potential in addressing these challenges. In this paper, we present an advanced automated progressive learning (AutoProg) framework for efficient training of LVMs. We begin by focusing on the pre-training of LVMs, using ViTs as a case study, and propose AutoProg-One, an AutoProg scheme featuring momentum growth (MoGrow) and a one-shot growth schedule search. Beyond pre-training, we extend our approach to tackle transfer learning and fine-tuning of LVMs. We expand the scope of AutoProg to cover a wider range of LVMs, including diffusion models. First, we introduce AutoProg-Zero, by enhancing the AutoProg framework with a novel zero-shot unfreezing schedule search, eliminating the need for one-shot supernet training. Second, we introduce a novel Unique Stage Identifier (SID) scheme to bridge the gap during network growth. These innovations, integrated with the core principles of AutoProg, offer a comprehensive solution for efficient training across various LVM scenarios. Extensive experiments show that AutoProg accelerates ViT pre-training by up to 1.85x on ImageNet and accelerates fine-tuning of diffusion models by up to 2.86x, with comparable or even higher performance. This work provides a robust and scalable approach to efficient training of LVMs, with potential applications in a wide range of vision tasks. Code: https://github.com/changlin31/AutoProg-Zero
Authors: Muyao Peng, Pei An, Zichen Wan, You Yang, Qiong Liu
Abstract: Along with the advancements in artificial intelligence technologies, image-to-point-cloud registration (I2P) techniques have made significant strides. Nevertheless, the dimensional differences in the features of points cloud (three-dimension) and image (two-dimension) continue to pose considerable challenges to their development. The primary challenge resides in the inability to leverage the features of one modality to augment those of another, thereby complicating the alignment of features within the latent space. To address this challenge, we propose an image-to-point-cloud method named as TFCT-I2P. Initially, we introduce a Three-Stream Fusion Network (TFN), which integrates color information from images with structural information from point clouds, facilitating the alignment of features from both modalities. Subsequently, to effectively mitigate patch-level misalignments introduced by the inclusion of color information, we design a Color-Aware Transformer (CAT). Finally, we conduct extensive experiments on 7Scenes, RGB-D Scenes V2, ScanNet V2, and a self-collected dataset. The results demonstrate that TFCT-I2P surpasses state-of-the-art methods by 1.5% in Inlier Ratio, 0.4% in Feature Matching Recall, and 5.4% in Registration Recall. Therefore, we believe that the proposed TFCT-I2P contributes to the advancement of I2P registration.
Authors: Riadul Islam, Sri Ranga Sai Krishna Tummala, Joey Mul\'e, Rohith Kankipati, Suraj Jalapally, Dhandeep Challagundla, Chad Howard, Ryan Robucci
Abstract: Smart focal-plane and in-chip image processing has emerged as a crucial technology for vision-enabled embedded systems with energy efficiency and privacy. However, the lack of special datasets providing examples of the data that these neuromorphic sensors compute to convey visual information has hindered the adoption of these promising technologies. Neuromorphic imager variants, including event-based sensors, produce various representations such as streams of pixel addresses representing time and locations of intensity changes in the focal plane, temporal-difference data, data sifted/thresholded by temporal differences, image data after applying spatial transformations, optical flow data, and/or statistical representations. To address the critical barrier to entry, we provide an annotated, temporal-threshold-based vision dataset specifically designed for face detection tasks derived from the same videos used for Aff-Wild2. By offering multiple threshold levels (e.g., 4, 8, 12, and 16), this dataset allows for comprehensive evaluation and optimization of state-of-the-art neural architectures under varying conditions and settings compared to traditional methods. The accompanying tool flow for generating event data from raw videos further enhances accessibility and usability. We anticipate that this resource will significantly support the development of robust vision systems based on smart sensors that can process based on temporal-difference thresholds, enabling more accurate and efficient object detection and localization and ultimately promoting the broader adoption of low-power, neuromorphic imaging technologies. To support further research, we publicly released the dataset at \url{https://dx.doi.org/10.21227/bw2e-dj78}.
Authors: Xiao Wang, Fuling Wang, Yuehang Li, Qingchuan Ma, Shiao Wang, Bo Jiang, Chuanfu Li, Jin Tang
Abstract: X-ray image-based medical report generation (MRG) is a pivotal area in artificial intelligence which can significantly reduce diagnostic burdens and patient wait times. Despite significant progress, we believe that the task has reached a bottleneck due to the limited benchmark datasets and the existing large models' insufficient capability enhancements in this specialized domain. Specifically, the recently released CheXpert Plus dataset lacks comparative evaluation algorithms and their results, providing only the dataset itself. This situation makes the training, evaluation, and comparison of subsequent algorithms challenging. Thus, we conduct a comprehensive benchmarking of existing mainstream X-ray report generation models and large language models (LLMs), on the CheXpert Plus dataset. We believe that the proposed benchmark can provide a solid comparative basis for subsequent algorithms and serve as a guide for researchers to quickly grasp the state-of-the-art models in this field. More importantly, we propose a large model for the X-ray image report generation using a multi-stage pre-training strategy, including self-supervised autoregressive generation and Xray-report contrastive learning, and supervised fine-tuning. Extensive experimental results indicate that the autoregressive pre-training based on Mamba effectively encodes X-ray images, and the image-text contrastive pre-training further aligns the feature spaces, achieving better experimental results. Source code can be found on \url{https://github.com/Event-AHU/Medical_Image_Analysis}.
Authors: Zaid Ilyas, Naveed Akhtar, David Suter, Syed Zulqarnain Gilani
Abstract: Image restoration and spectral reconstruction are longstanding computer vision tasks. Currently, CNN-transformer hybrid models provide state-of-the-art performance for these tasks. The key common ingredient in the architectural designs of these models is Channel-wise Self-Attention (CSA). We first show that CSA is an overall low-rank operation. Then, we propose an instance-Guided Low-rank Multi-Head selfattention (GLMHA) to replace the CSA for a considerable computational gain while closely retaining the original model performance. Unique to the proposed GLMHA is its ability to provide computational gain for both short and long input sequences. In particular, the gain is in terms of both Floating Point Operations (FLOPs) and parameter count reduction. This is in contrast to the existing popular computational complexity reduction techniques, e.g., Linformer, Performer, and Reformer, for whom FLOPs overpower the efficient design tricks for the shorter input sequences. Moreover, parameter reduction remains unaccounted for in the existing methods.We perform an extensive evaluation for the tasks of spectral reconstruction from RGB images, spectral reconstruction from snapshot compressive imaging, motion deblurring, and image deraining by enhancing the best-performing models with our GLMHA. Our results show up to a 7.7 Giga FLOPs reduction with 370K fewer parameters required to closely retain the original performance of the best-performing models that employ CSA.
Authors: Hongchao Shu, Mingxu Liu, Lalithkumar Seenivasan, Suxi Gu, Ping-Cheng Ku, Jonathan Knopf, Russell Taylor, Mathias Unberath
Abstract: Arthroscopy is a minimally invasive surgical procedure used to diagnose and treat joint problems. The clinical workflow of arthroscopy typically involves inserting an arthroscope into the joint through a small incision, during which surgeons navigate and operate largely by relying on their visual assessment through the arthroscope. However, the arthroscope's restricted field of view and lack of depth perception pose challenges in navigating complex articular structures and achieving surgical precision during procedures. Aiming at enhancing intraoperative awareness, we present a robust pipeline that incorporates simultaneous localization and mapping, depth estimation, and 3D Gaussian splatting to realistically reconstruct intra-articular structures solely based on monocular arthroscope video. Extending 3D reconstruction to Augmented Reality (AR) applications, our solution offers AR assistance for articular notch measurement and annotation anchoring in a human-in-the-loop manner. Compared to traditional Structure-from-Motion and Neural Radiance Field-based methods, our pipeline achieves dense 3D reconstruction and competitive rendering fidelity with explicit 3D representation in 7 minutes on average. When evaluated on four phantom datasets, our method achieves RMSE = 2.21mm reconstruction error, PSNR = 32.86 and SSIM = 0.89 on average. Because our pipeline enables AR reconstruction and guidance directly from monocular arthroscopy without any additional data and/or hardware, our solution may hold the potential for enhancing intraoperative awareness and facilitating surgical precision in arthroscopy. Our AR measurement tool achieves accuracy within 1.59 +/- 1.81mm and the AR annotation tool achieves a mIoU of 0.721.
Authors: Zhi Xu, Shaozhe Hao, Kai Han
Abstract: Enabling generative models to decompose visual concepts from a single image is a complex and challenging problem. In this paper, we study a new and challenging task, customized concept decomposition, wherein the objective is to leverage diffusion models to decompose a single image and generate visual concepts from various perspectives. To address this challenge, we propose a two-stage framework, CusConcept (short for Customized Visual Concept Decomposition), to extract customized visual concept embedding vectors that can be embedded into prompts for text-to-image generation. In the first stage, CusConcept employs a vocabulary-guided concept decomposition mechanism to build vocabularies along human-specified conceptual axes. The decomposed concepts are obtained by retrieving corresponding vocabularies and learning anchor weights. In the second stage, joint concept refinement is performed to enhance the fidelity and quality of generated images. We further curate an evaluation benchmark for assessing the performance of the open-world concept decomposition task. Our approach can effectively generate high-quality images of the decomposed concepts and produce related lexical predictions as secondary outcomes. Extensive qualitative and quantitative experiments demonstrate the effectiveness of CusConcept.
Authors: Mazen Balat, Mahmoud Essam Gabr, Hend Bakr, Ahmed B. Zaky
Abstract: The rise of short-form videos on platforms like TikTok has brought new challenges in safeguarding young viewers from inappropriate content. Traditional moderation methods often fall short in handling the vast and rapidly changing landscape of user-generated videos, increasing the risk of children encountering harmful material. This paper introduces TikGuard, a transformer-based deep learning approach aimed at detecting and flagging content unsuitable for children on TikTok. By using a specially curated dataset, TikHarm, and leveraging advanced video classification techniques, TikGuard achieves an accuracy of 86.7%, showing a notable improvement over existing methods in similar contexts. While direct comparisons are limited by the uniqueness of the TikHarm dataset, TikGuard's performance highlights its potential in enhancing content moderation, contributing to a safer online experience for minors. This study underscores the effectiveness of transformer models in video classification and sets a foundation for future research in this area.
Authors: Yunnan Wang, Ziqiang Li, Zequn Zhang, Wenyao Zhang, Baao Xie, Xihui Liu, Wenjun Zeng, Xin Jin
Abstract: There has been exciting progress in generating images from natural language or layout conditions. However, these methods struggle to faithfully reproduce complex scenes due to the insufficient modeling of multiple objects and their relationships. To address this issue, we leverage the scene graph, a powerful structured representation, for complex image generation. Different from the previous works that directly use scene graphs for generation, we employ the generative capabilities of variational autoencoders and diffusion models in a generalizable manner, compositing diverse disentangled visual clues from scene graphs. Specifically, we first propose a Semantics-Layout Variational AutoEncoder (SL-VAE) to jointly derive (layouts, semantics) from the input scene graph, which allows a more diverse and reasonable generation in a one-to-many mapping. We then develop a Compositional Masked Attention (CMA) integrated with a diffusion model, incorporating (layouts, semantics) with fine-grained attributes as generation guidance. To further achieve graph manipulation while keeping the visual content consistent, we introduce a Multi-Layered Sampler (MLS) for an "isolated" image editing effect. Extensive experiments demonstrate that our method outperforms recent competitors based on text, layout, or scene graph, in terms of generation rationality and controllability.
Authors: Hanqi Jiang, Xixuan Hao, Yuzhou Huang, Chong Ma, Jiaxun Zhang, Yi Pan, Ruimao Zhang
Abstract: This paper introduces an innovative approach to Medical Vision-Language Pre-training (Med-VLP) area in the specialized context of radiograph representation learning. While conventional methods frequently merge textual annotations into unified reports, we acknowledge the intrinsic hierarchical relationship between the findings and impression section in radiograph datasets. To establish a targeted correspondence between images and texts, we propose a novel HybridMED framework to align global-level visual representations with impression and token-level visual representations with findings. Moreover, our framework incorporates a generation decoder that employs two proxy tasks, responsible for generating the impression from (1) images, via a captioning branch, and (2) findings, through a summarization branch. Additionally, knowledge distillation is leveraged to facilitate the training process. Experiments on the MIMIC-CXR dataset reveal that our summarization branch effectively distills knowledge to the captioning branch, enhancing model performance without significantly increasing parameter requirements due to the shared self-attention and feed-forward architecture.
Authors: Bohong Chen, Yumeng Li, Yao-Xiang Ding, Tianjia Shao, Kun Zhou
Abstract: Current co-speech motion generation approaches usually focus on upper body gestures following speech contents only, while lacking supporting the elaborate control of synergistic full-body motion based on text prompts, such as talking while walking. The major challenges lie in 1) the existing speech-to-motion datasets only involve highly limited full-body motions, making a wide range of common human activities out of training distribution; 2) these datasets also lack annotated user prompts. To address these challenges, we propose SynTalker, which utilizes the off-the-shelf text-to-motion dataset as an auxiliary for supplementing the missing full-body motion and prompts. The core technical contributions are two-fold. One is the multi-stage training process which obtains an aligned embedding space of motion, speech, and prompts despite the significant distributional mismatch in motion between speech-to-motion and text-to-motion datasets. Another is the diffusion-based conditional inference process, which utilizes the separate-then-combine strategy to realize fine-grained control of local body parts. Extensive experiments are conducted to verify that our approach supports precise and flexible control of synergistic full-body motion generation based on both speeches and user prompts, which is beyond the ability of existing approaches.
Authors: Ivica Dimitrovski, Vlatko Spasev, Ivan Kitanovski
Abstract: Accurate semantic segmentation of remote sensing imagery is critical for various Earth observation applications, such as land cover mapping, urban planning, and environmental monitoring. However, individual data sources often present limitations for this task. Very High Resolution (VHR) aerial imagery provides rich spatial details but cannot capture temporal information about land cover changes. Conversely, Satellite Image Time Series (SITS) capture temporal dynamics, such as seasonal variations in vegetation, but with limited spatial resolution, making it difficult to distinguish fine-scale objects. This paper proposes a late fusion deep learning model (LF-DLM) for semantic segmentation that leverages the complementary strengths of both VHR aerial imagery and SITS. The proposed model consists of two independent deep learning branches. One branch integrates detailed textures from aerial imagery captured by UNetFormer with a Multi-Axis Vision Transformer (MaxViT) backbone. The other branch captures complex spatio-temporal dynamics from the Sentinel-2 satellite image time series using a U-Net with Temporal Attention Encoder (U-TAE). This approach leads to state-of-the-art results on the FLAIR dataset, a large-scale benchmark for land cover segmentation using multi-source optical imagery. The findings highlight the importance of multi-modality fusion in improving the accuracy and robustness of semantic segmentation in remote sensing applications.
Authors: Pranav Gupta, Advith Krishnan, Naman Nanda, Ananth Eswar, Deeksha Agarwal, Pratham Gohil, Pratyush Goel
Abstract: We present a novel dataset aimed at advancing danger analysis and assessment by addressing the challenge of quantifying danger in video content and identifying how human-like a Large Language Model (LLM) evaluator is for the same. This is achieved by compiling a collection of 100 YouTube videos featuring various events. Each video is annotated by human participants who provided danger ratings on a scale from 0 (no danger to humans) to 10 (life-threatening), with precise timestamps indicating moments of heightened danger. Additionally, we leverage LLMs to independently assess the danger levels in these videos using video summaries. We introduce Mean Squared Error (MSE) scores for multimodal meta-evaluation of the alignment between human and LLM danger assessments. Our dataset not only contributes a new resource for danger assessment in video content but also demonstrates the potential of LLMs in achieving human-like evaluations.
Authors: Rami Skaik, Leonardo Rossi, Tomaso Fontanini, Andrea Prati
Abstract: Recent advancements in generative models have revolutionized the field of artificial intelligence, enabling the creation of highly-realistic and detailed images. In this study, we propose a novel Mask Conditional Text-to-Image Generative Model (MCGM) that leverages the power of conditional diffusion models to generate pictures with specific poses. Our model builds upon the success of the Break-a-scene [1] model in generating new scenes using a single image with multiple subjects and incorporates a mask embedding injection that allows the conditioning of the generation process. By introducing this additional level of control, MCGM offers a flexible and intuitive approach for generating specific poses for one or more subjects learned from a single image, empowering users to influence the output based on their requirements. Through extensive experimentation and evaluation, we demonstrate the effectiveness of our proposed model in generating high-quality images that meet predefined mask conditions and improving the current Break-a-scene generative model.
Authors: Niki Maria Foteinopoulou, Enjie Ghorbel, Djamila Aouada
Abstract: Explainability in artificial intelligence is crucial for restoring trust, particularly in areas like face forgery detection, where viewers often struggle to distinguish between real and fabricated content. Vision and Large Language Models (VLLM) bridge computer vision and natural language, offering numerous applications driven by strong common-sense reasoning. Despite their success in various tasks, the potential of vision and language remains underexplored in face forgery detection, where they hold promise for enhancing explainability by leveraging the intrinsic reasoning capabilities of language to analyse fine-grained manipulation areas. As such, there is a need for a methodology that converts face forgery detection to a Visual Question Answering (VQA) task to systematically and fairly evaluate these capabilities. Previous efforts for unified benchmarks in deepfake detection have focused on the simpler binary task, overlooking evaluation protocols for fine-grained detection and text-generative models. We propose a multi-staged approach that diverges from the traditional binary decision paradigm to address this gap. In the first stage, we assess the models' performance on the binary task and their sensitivity to given instructions using several prompts. In the second stage, we delve deeper into fine-grained detection by identifying areas of manipulation in a multiple-choice VQA setting. In the third stage, we convert the fine-grained detection to an open-ended question and compare several matching strategies for the multi-label classification task. Finally, we qualitatively evaluate the fine-grained responses of the VLLMs included in the benchmark. We apply our benchmark to several popular models, providing a detailed comparison of binary, multiple-choice, and open-ended VQA evaluation across seven datasets. \url{https://nickyfot.github.io/hitchhickersguide.github.io/}
URLs: https://nickyfot.github.io/hitchhickersguide.github.io/
Authors: Dapeng Feng, Zhiqiang Chen, Yizhen Yin, Shipeng Zhong, Yuhua Qi, Hongbo Chen
Abstract: Simultaneous Localization and Mapping (SLAM) is pivotal in robotics, with photorealistic scene reconstruction emerging as a key challenge. To address this, we introduce Computational Alignment for Real-Time Gaussian Splatting SLAM (CaRtGS), a novel method enhancing the efficiency and quality of photorealistic scene reconstruction in real-time environments. Leveraging 3D Gaussian Splatting (3DGS), CaRtGS achieves superior rendering quality and processing speed, which is crucial for scene photorealistic reconstruction. Our approach tackles computational misalignment in Gaussian Splatting SLAM (GS-SLAM) through an adaptive strategy that optimizes training, addresses long-tail optimization, and refines densification. Experiments on Replica and TUM-RGBD datasets demonstrate CaRtGS's effectiveness in achieving high-fidelity rendering with fewer Gaussian primitives. This work propels SLAM towards real-time, photorealistic dense rendering, significantly advancing photorealistic scene representation. For the benefit of the research community, we release the code on our project website: https://dapengfeng.github.io/cartgs.
Authors: Yida Lin, Bing Xue, Mengjie Zhang, Sam Schofield, Richard Green
Abstract: This research focuses on the development of a drone equipped with pruning tools and a stereo vision camera to accurately detect and measure the spatial positions of tree branches. YOLO is employed for branch segmentation, while two depth estimation approaches, monocular and stereo, are investigated. In comparison to SGBM, deep learning techniques produce more refined and accurate depth maps. In the absence of ground-truth data, a fine-tuning process using deep neural networks is applied to approximate optimal depth values. This methodology facilitates precise branch detection and distance measurement, addressing critical challenges in the automation of pruning operations. The results demonstrate notable advancements in both accuracy and efficiency, underscoring the potential of deep learning to drive innovation and enhance automation in the agricultural sector.
Authors: Alberto Presta, Enzo Tartaglione, Attilio Fiandrotti, Marco Grangetto
Abstract: In end-to-end learned image compression, encoder and decoder are jointly trained to minimize a $R + {\lambda}D$ cost function, where ${\lambda}$ controls the trade-off between rate of the quantized latent representation and image quality. Unfortunately, a distinct encoder-decoder pair with millions of parameters must be trained for each ${\lambda}$, hence the need to switch encoders and to store multiple encoders and decoders on the user device for every target rate. This paper proposes to exploit a differentiable quantizer designed around a parametric sum of hyperbolic tangents, called STanH , that relaxes the step-wise quantization function. STanH is implemented as a differentiable activation layer with learnable quantization parameters that can be plugged into a pre-trained fixed rate model and refined to achieve different target bitrates. Experimental results show that our method enables variable rate coding with comparable efficiency to the state-of-the-art, yet with significant savings in terms of ease of deployment, training time, and storage costs
Authors: Aurora Micheli, Olaf Booij, Jan van Gemert, Nergis T\"omen
Abstract: Spiking Neural Networks (SNNs) and neuromorphic computing offer bio-inspired advantages such as sparsity and ultra-low power consumption, providing a promising alternative to conventional networks. However, training deep SNNs from scratch remains a challenge, as SNNs process and transmit information by quantizing the real-valued membrane potentials into binary spikes. This can lead to information loss and vanishing spikes in deeper layers, impeding effective training. While weight initialization is known to be critical for training deep neural networks, what constitutes an effective initial state for a deep SNN is not well-understood. Existing weight initialization methods designed for conventional networks (ANNs) are often applied to SNNs without accounting for their distinct computational properties. In this work we derive an optimal weight initialization method specifically tailored for SNNs, taking into account the quantization operation. We show theoretically that, unlike standard approaches, this method enables the propagation of activity in deep SNNs without loss of spikes. We demonstrate this behavior in numerical simulations of SNNs with up to 100 layers across multiple time steps. We present an in-depth analysis of the numerical conditions, regarding layer width and neuron hyperparameters, which are necessary to accurately apply our theoretical findings. Furthermore, our experiments on MNIST demonstrate higher accuracy and faster convergence when using the proposed weight initialization scheme. Finally, we show that the newly introduced weight initialization is robust against variations in several network and neuron hyperparameters.
Authors: Pengxi Zeng, Alberto Presta, Jonah Reinis, Dinesh Bharadia, Hang Qiu, Pamela Cosman
Abstract: Efficient point cloud (PC) compression is crucial for streaming applications, such as augmented reality and cooperative perception. Classic PC compression techniques encode all the points in a frame. Tailoring compression towards perception tasks at the receiver side, we ask the question, "Can we remove the ground points during transmission without sacrificing the detection performance?" Our study reveals a strong dependency on the ground from state-of-the-art (SOTA) 3D object detection models, especially on those points below and around the object. In this work, we propose a lightweight obstacle-aware Pillar-based Ground Removal (PGR) algorithm. PGR filters out ground points that do not provide context to object recognition, significantly improving compression ratio without sacrificing the receiver side perception performance. Not using heavy object detection or semantic segmentation models, PGR is light-weight, highly parallelizable, and effective. Our evaluations on KITTI and Waymo Open Dataset show that SOTA detection models work equally well with PGR removing 20-30% of the points, with a speeding of 86 FPS.
Authors: Geng Li, Haozhi Cao, Mingyang Liu, Shenghai Yuan, Jianfei Yang
Abstract: Point cloud registration aims to provide estimated transformations to align point clouds, which plays a crucial role in pose estimation of various navigation systems, such as surgical guidance systems and autonomous vehicles. Despite the impressive performance of recent models on benchmark datasets, many rely on complex modules like KPConv and Transformers, which impose significant computational and memory demands. These requirements hinder their practical application, particularly in resource-constrained environments such as mobile robotics. In this paper, we propose a novel point cloud registration network that leverages a pure MLP architecture, constructing geometric information offline. This approach eliminates the computational and memory burdens associated with traditional complex feature extractors and significantly reduces inference time and resource consumption. Our method is the first to replace 3D coordinate inputs with offline-constructed geometric encoding, improving generalization and stability, as demonstrated by Maximum Mean Discrepancy (MMD) comparisons. This efficient and accurate geometric representation marks a significant advancement in point cloud analysis, particularly for applications requiring fast and reliability.
Authors: Shunyi Zhao, Zehuan Yu, Zuxin Fan, Zhihao Zhou, Lecheng Ruan, Qining Wang
Abstract: Visual features, whose description often relies on the local intensity and gradient direction, have found wide applications in robot navigation and localization in recent years. However, the extraction of visual features is usually disturbed by the variation of illumination conditions, making it challenging for real-world applications. Previous works have addressed this issue by establishing datasets with variations in illumination conditions, but can be costly and time-consuming. This paper proposes a design procedure for an illumination-robust feature extractor, where the recently developed relightable 3D reconstruction techniques are adopted for rapid and direct data generation with varying illumination conditions. A self-supervised framework is proposed for extracting features with advantages in repeatability for key points and similarity for descriptors across good and bad illumination conditions. Experiments are conducted to demonstrate the effectiveness of the proposed method for robust feature extraction. Ablation studies also indicate the effectiveness of the self-supervised framework design.
Authors: Marcel C. B\"uhler, Gengyan Li, Erroll Wood, Leonhard Helminger, Xu Chen, Tanmay Shah, Daoye Wang, Stephan Garbin, Sergio Orts-Escolano, Otmar Hilliges, Dmitry Lagun, J\'er\'emy Riviere, Paulo Gotardo, Thabo Beeler, Abhimitra Meka, Kripasindhu Sarkar
Abstract: Volumetric modeling and neural radiance field representations have revolutionized 3D face capture and photorealistic novel view synthesis. However, these methods often require hundreds of multi-view input images and are thus inapplicable to cases with less than a handful of inputs. We present a novel volumetric prior on human faces that allows for high-fidelity expressive face modeling from as few as three input views captured in the wild. Our key insight is that an implicit prior trained on synthetic data alone can generalize to extremely challenging real-world identities and expressions and render novel views with fine idiosyncratic details like wrinkles and eyelashes. We leverage a 3D Morphable Face Model to synthesize a large training set, rendering each identity with different expressions, hair, clothing, and other assets. We then train a conditional Neural Radiance Field prior on this synthetic dataset and, at inference time, fine-tune the model on a very sparse set of real images of a single subject. On average, the fine-tuning requires only three inputs to cross the synthetic-to-real domain gap. The resulting personalized 3D model reconstructs strong idiosyncratic facial expressions and outperforms the state-of-the-art in high-quality novel view synthesis of faces from sparse inputs in terms of perceptual and photo-metric quality.
Authors: {\DJ}or{\dj}e Nedeljkovi\'c
Abstract: Cross-camera data association is one of the cornerstones of the multi-camera computer vision field. Although often integrated into detection and tracking tasks through architecture design and loss definition, it is also recognized as an independent challenge. The ultimate goal is to connect appearances of one item from all cameras, wherever it is visible. Therefore, one possible perspective on this task involves supervised clustering of the affinity graph, where nodes are instances captured by all cameras. They are represented by appropriate visual features and positional attributes. We leverage the advantages of GNN (Graph Neural Network) architecture to examine nodes' relations and generate representative edge embeddings. These embeddings are then classified to determine the existence or non-existence of connections in node pairs. Therefore, the core of this approach is graph connectivity prediction. Experimental validation was conducted on multicamera pedestrian datasets across diverse environments such as the laboratory, basketball court, and terrace. Our proposed method, named SGC-CCA, outperformed the state-of-the-art method named GNN-CCA across all clustering metrics, offering an end-to-end clustering solution without the need for graph post-processing. The code is available at https://github.com/djordjened92/cca-gnnclust.
Authors: Youngho Yoon, Hyun-Kurl Jang, Kuk-Jin Yoon
Abstract: Novel view synthesis (NVS) aims to generate images at arbitrary viewpoints using multi-view images, and recent insights from neural radiance fields (NeRF) have contributed to remarkable improvements. Recently, studies on generalizable NeRF (G-NeRF) have addressed the challenge of per-scene optimization in NeRFs. The construction of radiance fields on-the-fly in G-NeRF simplifies the NVS process, making it well-suited for real-world applications. Meanwhile, G-NeRF still struggles in representing fine details for a specific scene due to the absence of per-scene optimization, even with texture-rich multi-view source inputs. As a remedy, we propose a Geometry-driven Multi-reference Texture transfer network (GMT) available as a plug-and-play module designed for G-NeRF. Specifically, we propose ray-imposed deformable convolution (RayDCN), which aligns input and reference features reflecting scene geometry. Additionally, the proposed texture preserving transformer (TP-Former) aggregates multi-view source features while preserving texture information. Consequently, our module enables direct interaction between adjacent pixels during the image enhancement process, which is deficient in G-NeRF models with an independent rendering process per pixel. This addresses constraints that hinder the ability to capture high-frequency details. Experiments show that our plug-and-play module consistently improves G-NeRF models on various benchmark datasets.
Authors: Mazen Balat, Rewaa Awaad, Hend Adel, Ahmed B. Zaky, Salah A. Aly
Abstract: This paper presents an Arabic Alphabet Sign Language recognition approach, using deep learning methods in conjunction with transfer learning and transformer-based models. We study the performance of the different variants on two publicly available datasets, namely ArSL2018 and AASL. This task will make full use of state-of-the-art CNN architectures like ResNet50, MobileNetV2, and EfficientNetB7, and the latest transformer models such as Google ViT and Microsoft Swin Transformer. These pre-trained models have been fine-tuned on the above datasets in an attempt to capture some unique features of Arabic sign language motions. Experimental results present evidence that the suggested methodology can receive a high recognition accuracy, by up to 99.6\% and 99.43\% on ArSL2018 and AASL, respectively. That is far beyond the previously reported state-of-the-art approaches. This performance opens up even more avenues for communication that may be more accessible to Arabic-speaking deaf and hard-of-hearing, and thus encourages an inclusive society.
Authors: Saurav Jha, Shiqi Yang, Masato Ishii, Mengjie Zhao, Christian Simon, Jehanzeb Mirza, Dong Gong, Lina Yao, Shusuke Takahashi, Yuki Mitsufuji
Abstract: Personalized text-to-image diffusion models have grown popular for their ability to efficiently acquire a new concept from user-defined text descriptions and a few images. However, in the real world, a user may wish to personalize a model on multiple concepts but one at a time, with no access to the data from previous concepts due to storage/privacy concerns. When faced with this continual learning (CL) setup, most personalization methods fail to find a balance between acquiring new concepts and retaining previous ones -- a challenge that continual personalization (CP) aims to solve. Inspired by the successful CL methods that rely on class-specific information for regularization, we resort to the inherent class-conditioned density estimates, also known as diffusion classifier (DC) scores, for continual personalization of text-to-image diffusion models. Namely, we propose using DC scores for regularizing the parameter-space and function-space of text-to-image diffusion models, to achieve continual personalization. Using several diverse evaluation setups, datasets, and metrics, we show that our proposed regularization-based CP methods outperform the state-of-the-art C-LoRA, and other baselines. Finally, by operating in the replay-free CL setup and on low-rank adapters, our method incurs zero storage and parameter overhead, respectively, over the state-of-the-art.
Authors: Raktim Gautam Goswami, Naman Patel, Prashanth Krishnamurthy, Farshad Khorrami
Abstract: Map-free LiDAR localization systems accurately localize within known environments by predicting sensor position and orientation directly from raw point clouds, eliminating the need for large maps and descriptors. However, their long training times hinder rapid adaptation to new environments. To address this, we propose FlashMix, which uses a frozen, scene-agnostic backbone to extract local point descriptors, aggregated with an MLP mixer to predict sensor pose. A buffer of local descriptors is used to accelerate training by orders of magnitude, combined with metric learning or contrastive loss regularization of aggregated descriptors to improve performance and convergence. We evaluate FlashMix on various LiDAR localization benchmarks, examining different regularizations and aggregators, demonstrating its effectiveness for rapid and accurate LiDAR localization in real-world scenarios. The code is available at https://github.com/raktimgg/FlashMix.
Authors: \'Alvaro Heredia-Lid\'on, Luis M. Echeverry-Quiceno, Alejandro Gonz\'alez, Noem\'i Hostalet, Edith Pomarol-Clotet, Juan Fortea, Mar Fatj\'o-Vilas, Neus Mart\'inez-Abad\'ias, Xavier Sevillano
Abstract: Facial dysmorphologies have emerged as potential critical indicators in the diagnosis and prognosis of genetic, psychotic and rare disorders. While in certain conditions these dysmorphologies are severe, in other cases may be subtle and not perceivable to the human eye, requiring precise quantitative tools for their identification. Manual coding of facial dysmorphologies is a burdensome task and is subject to inter- and intra-observer variability. To overcome this gap, we present BioFace3D as a fully automatic tool for the calculation of facial biomarkers using facial models reconstructed from magnetic resonance images. The tool is divided into three automatic modules for the extraction of 3D facial models from magnetic resonance images, the registration of homologous 3D landmarks encoding facial morphology, and the calculation of facial biomarkers from anatomical landmarks coordinates using geometric morphometrics techniques.
Authors: Kaichen Zhou, Yang Cao, Teawhan Kim, Hao Zhao, Hao Dong, Kai Ming Ting, Ye Zhu
Abstract: Recent advancements in industrial anomaly detection have been hindered by the lack of realistic datasets that accurately represent real-world conditions. Existing algorithms are often developed and evaluated using idealized datasets, which deviate significantly from real-life scenarios characterized by environmental noise and data corruption such as fluctuating lighting conditions, variable object poses, and unstable camera positions. To address this gap, we introduce the Realistic Anomaly Detection (RAD) dataset, the first multi-view RGB-based anomaly detection dataset specifically collected using a real robot arm, providing unique and realistic data scenarios. RAD comprises 4765 images across 13 categories and 4 defect types, collected from more than 50 viewpoints, providing a comprehensive and realistic benchmark. This multi-viewpoint setup mirrors real-world conditions where anomalies may not be detectable from every perspective. Moreover, by sampling varying numbers of views, the algorithm's performance can be comprehensively evaluated across different viewpoints. This approach enhances the thoroughness of performance assessment and helps improve the algorithm's robustness. Besides, to support 3D multi-view reconstruction algorithms, we propose a data augmentation method to improve the accuracy of pose estimation and facilitate the reconstruction of 3D point clouds. We systematically evaluate state-of-the-art RGB-based and point cloud-based models using RAD, identifying limitations and future research directions. The code and dataset could found at https://github.com/kaichen-z/RAD
Authors: Vihang Patil, Andreas Radler, Daniel Klotz, Sepp Hochreiter
Abstract: Humans excel at abstracting data and constructing \emph{reusable} concepts, a capability lacking in current continual learning systems. The field of object-centric learning addresses this by developing abstract representations, or slots, from data without human supervision. Different methods have been proposed to tackle this task for images, whereas most are overly complex, non-differentiable, or poorly scalable. In this paper, we introduce a conceptually simple, fully-differentiable, non-iterative, and scalable method called SAMP Simplified Slot Attention with Max Pool Priors). It is implementable using only Convolution and MaxPool layers and an Attention layer. Our method encodes the input image with a Convolutional Neural Network and then uses a branch of alternating Convolution and MaxPool layers to create specialized sub-networks and extract primitive slots. These primitive slots are then used as queries for a Simplified Slot Attention over the encoded image. Despite its simplicity, our method is competitive or outperforms previous methods on standard benchmarks.
Authors: Lakshmi Nair
Abstract: Synthetic data generation is an important application of machine learning in the field of medical imaging. While existing approaches have successfully applied fine-tuned diffusion models for synthesizing medical images, we explore potential improvements to this pipeline through feature-aligned diffusion. Our approach aligns intermediate features of the diffusion model to the output features of an expert, and our preliminary findings show an improvement of 9% in generation accuracy and ~0.12 in SSIM diversity. Our approach is also synergistic with existing methods, and easily integrated into diffusion training pipelines for improvements. We make our code available at \url{https://github.com/lnairGT/Feature-Aligned-Diffusion}.
Authors: Robert Krajewski, Huijo Kim
Abstract: High-definition maps (HD maps) play a crucial role in the development, safety validation, and operation of highly automated vehicles. Efficiently collecting up-to-date sensor data from road segments and obtaining accurate maps from these are key challenges in HD map creation. Commonly used methods, such as dedicated measurement vehicles and crowd-sourced data from series vehicles, often face limitations in commercial viability. Although high-resolution aerial imagery offers a cost-effective or even free alternative, it requires significant manual effort and time to transform it into maps. In this paper, we introduce a semi-automatic method for creating HD maps from high-resolution aerial imagery. Our method involves training neural networks to semantically segment aerial images into classes relevant to HD maps. The resulting segmentation is then hierarchically post-processed to generate a prototypical HD map of visible road elements. Exporting the map to the Lanelet2 format allows easy extension for different use cases using standard tools. To train and evaluate our method, we created a dataset using public aerial imagery of urban road segments in Germany. In our evaluation, we achieved an automatic mapping of lane markings and road borders with a recall and precision exceeding 96%. The source code for our method is publicly available at https://github.com/RobertKrajewski/DeepAerialMapper.
Authors: Chen Cai, Zheng Wang, Jianjun Gao, Wenyang Liu, Ye Lu, Runzhong Zhang, Kim-Hui Yap
Abstract: In recent years, the rapid increase in online video content has underscored the limitations of static Video Question Answering (VideoQA) models trained on fixed datasets, as they struggle to adapt to new questions or tasks posed by newly available content. In this paper, we explore the novel challenge of VideoQA within a continual learning framework, and empirically identify a critical issue: fine-tuning a large language model (LLM) for a sequence of tasks often results in catastrophic forgetting. To address this, we propose Collaborative Prompting (ColPro), which integrates specific question constraint prompting, knowledge acquisition prompting, and visual temporal awareness prompting. These prompts aim to capture textual question context, visual content, and video temporal dynamics in VideoQA, a perspective underexplored in prior research. Experimental results on the NExT-QA and DramaQA datasets show that ColPro achieves superior performance compared to existing approaches, achieving 55.14\% accuracy on NExT-QA and 71.24\% accuracy on DramaQA, highlighting its practical relevance and effectiveness.
Authors: Wenwen Qiang, Zeen Song, Ziyin Gu, Jiangmeng Li, Changwen Zheng, Fuchun Sun, Hui Xiong
Abstract: Self-supervised learning (SSL) methods learn from unlabeled data and achieve high generalization performance on downstream tasks. However, they may also suffer from overfitting to their training data and lose the ability to adapt to new tasks. To investigate this phenomenon, we conduct experiments on various SSL methods and datasets and make two observations: (1) Overfitting occurs abruptly in later layers and epochs, while generalizing features are learned in early layers for all epochs; (2) Coding rate reduction can be used as an indicator to measure the degree of overfitting in SSL models. Based on these observations, we propose Undoing Memorization Mechanism (UMM), a plug-and-play method that mitigates overfitting of the pre-trained feature extractor by aligning the feature distributions of the early and the last layers to maximize the coding rate reduction of the last layer output. The learning process of UMM is a bi-level optimization process. We provide a causal analysis of UMM to explain how UMM can help the pre-trained feature extractor overcome overfitting and recover generalization. We also demonstrate that UMM significantly improves the generalization performance of SSL methods on various downstream tasks.
Authors: Mostafa Hajighasemloua, Samad Sheikhaei, Hamid Soltanian-Zadeha
Abstract: Artificial intelligence algorithms have demonstrated their image classification and segmentation ability in the past decade. However, artificial intelligence algorithms perform less for actual clinical data than those used for simulations. This research aims to present a novel hybrid learning model using self-supervised learning and knowledge distillation, which can achieve sufficient generalization and robustness. The self-attention mechanism and tokens employed in ViT, besides the local-to-global learning approach used in the hybrid model, enable the proposed algorithm to extract a high-dimensional and high-quality feature space from images. To demonstrate the proposed neural network's capability in classifying and extracting feature spaces from medical images, we use it on a dataset of Diabetic Retinopathy images, specifically the EyePACS dataset. This dataset is more complex structurally and challenging regarding damaged areas than other medical images. For the first time in this study, self-supervised learning and knowledge distillation are used to classify this dataset. In our algorithm, for the first time among all self-supervised learning and knowledge distillation models, the test dataset is 50% larger than the training dataset. Unlike many studies, we have not removed any images from the dataset. Finally, our algorithm achieved an accuracy of 79.1% in the linear classifier and 74.36% in the k-NN algorithm for multiclass classification. Compared to a similar state-of-the-art model, our results achieved higher accuracy and more effective representation spaces.
Authors: Gabriele Spadaro, Marco Grangetto, Attilio Fiandrotti, Enzo Tartaglione, Jhony H. Giraldo
Abstract: In recent years, Graph Neural Networks (GNNs) have demonstrated strong adaptability to various real-world challenges, with architectures such as Vision GNN (ViG) achieving state-of-the-art performance in several computer vision tasks. However, their practical applicability is hindered by the computational complexity of constructing the graph, which scales quadratically with the image size. In this paper, we introduce a novel Windowed vision Graph neural Network (WiGNet) model for efficient image processing. WiGNet explores a different strategy from previous works by partitioning the image into windows and constructing a graph within each window. Therefore, our model uses graph convolutions instead of the typical 2D convolution or self-attention mechanism. WiGNet effectively manages computational and memory complexity for large image sizes. We evaluate our method in the ImageNet-1k benchmark dataset and test the adaptability of WiGNet using the CelebA-HQ dataset as a downstream task with higher-resolution images. In both of these scenarios, our method achieves competitive results compared to previous vision GNNs while keeping memory and computational complexity at bay. WiGNet offers a promising solution toward the deployment of vision GNNs in real-world applications. We publicly released the code at https://github.com/EIDOSLAB/WiGNet.
Authors: Rinor Cakaj, Jens Mehnert, Bin Yang
Abstract: Convolutional Neural Networks (CNNs) are important for many machine learning tasks. They are built with different types of layers: convolutional layers that detect features, dropout layers that help to avoid over-reliance on any single neuron, and residual layers that allow the reuse of features. However, CNNs lack a dynamic feature retention mechanism similar to the human brain's memory, limiting their ability to use learned information in new contexts. To bridge this gap, we introduce the "Squeeze-and-Remember" (SR) block, a novel architectural unit that gives CNNs dynamic memory-like functionalities. The SR block selectively memorizes important features during training, and then adaptively re-applies these features during inference. This improves the network's ability to make contextually informed predictions. Empirical results on ImageNet and Cityscapes datasets demonstrate the SR block's efficacy: integration into ResNet50 improved top-1 validation accuracy on ImageNet by 0.52% over dropout2d alone, and its application in DeepLab v3 increased mean Intersection over Union in Cityscapes by 0.20%. These improvements are achieved with minimal computational overhead. This show the SR block's potential to enhance the capabilities of CNNs in image processing tasks.
Authors: Yunze Liu, Li Yi
Abstract: Mamba has achieved significant advantages in long-context modeling and autoregressive tasks, but its scalability with large parameters remains a major limitation in vision applications. pretraining is a widely used strategy to enhance backbone model performance. Although the success of Masked Autoencoder in Transformer pretraining is well recognized, it does not significantly improve Mamba's visual learning performance. We found that using the correct autoregressive pretraining can significantly boost the performance of the Mamba architecture. Based on this analysis, we propose Masked Autoregressive Pretraining (MAP) to pretrain a hybrid Mamba-Transformer vision backbone network. This strategy combines the strengths of both MAE and Autoregressive pretraining, improving the performance of Mamba and Transformer modules within a unified paradigm. Additionally, in terms of integrating Mamba and Transformer modules, we empirically found that inserting Transformer layers at regular intervals within Mamba layers can significantly enhance downstream task performance. Experimental results show that both the pure Mamba architecture and the hybrid Mamba-Transformer vision backbone network pretrained with MAP significantly outperform other pretraining strategies, achieving state-of-the-art performance. We validate the effectiveness of the method on both 2D and 3D datasets and provide detailed ablation studies to support the design choices for each component.
Authors: Junlin Han, Jianyuan Wang, Andrea Vedaldi, Philip Torr, Filippos Kokkinos
Abstract: Generating high-quality 3D content from text, single images, or sparse view images remains a challenging task with broad applications.Existing methods typically employ multi-view diffusion models to synthesize multi-view images, followed by a feed-forward process for 3D reconstruction. However, these approaches are often constrained by a small and fixed number of input views, limiting their ability to capture diverse viewpoints and, even worse, leading to suboptimal generation results if the synthesized views are of poor quality. To address these limitations, we propose Flex3D, a novel two-stage framework capable of leveraging an arbitrary number of high-quality input views. The first stage consists of a candidate view generation and curation pipeline. We employ a fine-tuned multi-view image diffusion model and a video diffusion model to generate a pool of candidate views, enabling a rich representation of the target 3D object. Subsequently, a view selection pipeline filters these views based on quality and consistency, ensuring that only the high-quality and reliable views are used for reconstruction. In the second stage, the curated views are fed into a Flexible Reconstruction Model (FlexRM), built upon a transformer architecture that can effectively process an arbitrary number of inputs. FlemRM directly outputs 3D Gaussian points leveraging a tri-plane representation, enabling efficient and detailed 3D generation. Through extensive exploration of design and training strategies, we optimize FlexRM to achieve superior performance in both reconstruction and generation tasks. Our results demonstrate that Flex3D achieves state-of-the-art performance, with a user study winning rate of over 92% in 3D generation tasks when compared to several of the latest feed-forward 3D generative models.
Authors: Robin Gerster, Holger Caesar, Matthias Rapp, Alexander Wolpert, Michael Teutsch
Abstract: Despite their success in various vision tasks, deep neural network architectures often underperform in out-of-distribution scenarios due to the difference between training and target domain style. To address this limitation, we introduce One-Shot Style Adaptation (OSSA), a novel unsupervised domain adaptation method for object detection that utilizes a single, unlabeled target image to approximate the target domain style. Specifically, OSSA generates diverse target styles by perturbing the style statistics derived from a single target image and then applies these styles to a labeled source dataset at the feature level using Adaptive Instance Normalization (AdaIN). Extensive experiments show that OSSA establishes a new state-of-the-art among one-shot domain adaptation methods by a significant margin, and in some cases, even outperforms strong baselines that use thousands of unlabeled target images. By applying OSSA in various scenarios, including weather, simulated-to-real (sim2real), and visual-to-thermal adaptations, our study explores the overarching significance of the style gap in these contexts. OSSA's simplicity and efficiency allow easy integration into existing frameworks, providing a potentially viable solution for practical applications with limited data availability. Code is available at https://github.com/RobinGerster7/OSSA
Authors: Yuheng Li, Haotian Liu, Mu Cai, Yijun Li, Eli Shechtman, Zhe Lin, Yong Jae Lee, Krishna Kumar Singh
Abstract: In this paper, we introduce a model designed to improve the prediction of image-text alignment, targeting the challenge of compositional understanding in current visual-language models. Our approach focuses on generating high-quality training datasets for the alignment task by producing mixed-type negative captions derived from positive ones. Critically, we address the distribution imbalance between positive and negative captions to ensure that the alignment model does not depend solely on textual information but also considers the associated images for predicting alignment accurately. By creating this enhanced training data, we fine-tune an existing leading visual-language model to boost its capability in understanding alignment. Our model significantly outperforms current top-performing methods across various datasets. We also demonstrate the applicability of our model by ranking the images generated by text-to-image models based on text alignment. Project page: \url{https://yuheng-li.github.io/LLaVA-score/}
Authors: Da-Wei Zhou, Zi-Wen Cai, Han-Jia Ye, Lijun Zhang, De-Chuan Zhan
Abstract: Domain-Incremental Learning (DIL) involves the progressive adaptation of a model to new concepts across different domains. While recent advances in pre-trained models provide a solid foundation for DIL, learning new concepts often results in the catastrophic forgetting of pre-trained knowledge. Specifically, sequential model updates can overwrite both the representation and the classifier with knowledge from the latest domain. Thus, it is crucial to develop a representation and corresponding classifier that accommodate all seen domains throughout the learning process. To this end, we propose DUal ConsolidaTion (Duct) to unify and consolidate historical knowledge at both the representation and classifier levels. By merging the backbone of different stages, we create a representation space suitable for multiple domains incrementally. The merged representation serves as a balanced intermediary that captures task-specific features from all seen domains. Additionally, to address the mismatch between consolidated embeddings and the classifier, we introduce an extra classifier consolidation process. Leveraging class-wise semantic information, we estimate the classifier weights of old domains within the latest embedding space. By merging historical and estimated classifiers, we align them with the consolidated embedding space, facilitating incremental classification. Extensive experimental results on four benchmark datasets demonstrate Duct's state-of-the-art performance.
Authors: Yujin Oh, Sangjoon Park, Xiang Li, Wang Yi, Jonathan Paly, Jason Efstathiou, Annie Chan, Jun Won Kim, Hwa Kyung Byun, Ik Jae Lee, Jaeho Cho, Chan Woo Wee, Peng Shu, Peilong Wang, Nathan Yu, Jason Holmes, Jong Chul Ye, Quanzheng Li, Wei Liu, Woong Sub Koom, Jin Sung Kim, Kyungsang Kim
Abstract: Clinical experts employ diverse philosophies and strategies in patient care, influenced by regional patient populations. However, existing medical artificial intelligence (AI) models are often trained on data distributions that disproportionately reflect highly prevalent patterns, reinforcing biases and overlooking the diverse expertise of clinicians. To overcome this limitation, we introduce the Mixture of Multicenter Experts (MoME) approach. This method strategically integrates specialized expertise from diverse clinical strategies, enhancing the AI model's ability to generalize and adapt across multiple medical centers. The MoME-based multimodal target volume delineation model, trained with few-shot samples including images and clinical notes from each medical center, outperformed baseline methods in prostate cancer radiotherapy target delineation. The advantages of MoME were most pronounced when data characteristics varied across centers or when data availability was limited, demonstrating its potential for broader clinical applications.Therefore, the MoME framework enables the deployment of AI-based target volume delineation models in resource-constrained medical facilities by adapting to specific preferences of each medical center only using a few sample data, without the need for data sharing between institutions. Expanding the number of multicenter experts within the MoME framework will significantly enhance the generalizability, while also improving the usability and adaptability of clinical AI applications in the field of precision radiation oncology.
Authors: Federico Fontana, Romeo Lanzino, Anxhelo Diko, Gian Luca Foresti, Luigi Cinque
Abstract: This paper works on Binary Neural Networks (BNNs), a promising avenue for efficient deep learning, offering significant reductions in computational overhead and memory footprint to full precision networks. However, the challenge of energy-intensive training and the drop in performance have been persistent issues. Tackling the challenge, prior works focus primarily on task-related inference optimization. Unlike prior works, this study offers an innovative methodology integrating BNNs with cyclic precision training, introducing the CycleBNN. This approach is designed to enhance training efficiency while minimizing the loss in performance. By dynamically adjusting precision in cycles, we achieve a convenient trade-off between training efficiency and model performance. This emphasizes the potential of our method in energy-constrained training scenarios, where data is collected onboard and paves the way for sustainable and efficient deep learning architectures. To gather insights on CycleBNN's efficiency, we conduct experiments on ImageNet, CIFAR-10, and PASCAL-VOC, obtaining competitive performances while using 96.09\% less operations during training on ImageNet, 88.88\% on CIFAR-10 and 96.09\% on PASCAL-VOC. Finally, CycleBNN offers a path towards faster, more accessible training of efficient networks, accelerating the development of practical applications. The PyTorch code is available at \url{https://github.com/fedeloper/CycleBNN/}
Authors: Haoran Li, Zhennan Jiang, Yuhui Chen, Dongbin Zhao
Abstract: With high-dimensional state spaces, visual reinforcement learning (RL) faces significant challenges in exploitation and exploration, resulting in low sample efficiency and training stability. As a time-efficient diffusion model, although consistency models have been validated in online state-based RL, it is still an open question whether it can be extended to visual RL. In this paper, we investigate the impact of non-stationary distribution and the actor-critic framework on consistency policy in online RL, and find that consistency policy was unstable during the training, especially in visual RL with the high-dimensional state space. To this end, we suggest sample-based entropy regularization to stabilize the policy training, and propose a consistency policy with prioritized proximal experience regularization (CP3ER) to improve sample efficiency. CP3ER achieves new state-of-the-art (SOTA) performance in 21 tasks across DeepMind control suite and Meta-world. To our knowledge, CP3ER is the first method to apply diffusion/consistency models to visual RL and demonstrates the potential of consistency models in visual RL. More visualization results are available at https://jzndd.github.io/CP3ER-Page/.
Authors: Chaohui Xu, Qi Cui, Jinxin Dong, Weiyang He, Chip-Hong Chang
Abstract: Illegitimate reproduction, distribution and derivation of Deep Neural Network (DNN) models can inflict economic loss, reputation damage and even privacy infringement. Passive DNN intellectual property (IP) protection methods such as watermarking and fingerprinting attempt to prove the ownership upon IP violation, but they are often too late to stop catastrophic damage of IP abuse and too feeble against strong adversaries. In this paper, we propose IDEA, an Inverse Domain Expert Adaptation based proactive DNN IP protection method featuring active authorization and source traceability. IDEA generalizes active authorization as an inverse problem of domain adaptation. The multi-adaptive optimization is solved by a mixture-of-experts model with one real and two fake experts. The real expert re-optimizes the source model to correctly classify test images with a unique model user key steganographically embedded. The fake experts are trained to output random prediction on test images without or with incorrect user key embedded by minimizing their mutual information (MI) with the real expert. The MoE model is knowledge distilled into a unified protected model to avoid leaking the expert model features by maximizing their MI with additional multi-layer attention and contrastive representation loss optimization. IDEA not only prevents unauthorized users without the valid key to access the functional model, but also enable the model owner to validate the deployed model and trace the source of IP infringement. We extensively evaluate IDEA on five datasets and four DNN models to demonstrate its effectiveness in authorization control, culprit tracing success rate, and robustness against various attacks.
Authors: Aymane Khaldi, El Mostafa Kalmoun
Abstract: Pumpkin is a vital crop cultivated globally, and its productivity is crucial for food security, especially in developing regions. Accurate and timely detection of pumpkin leaf diseases is essential to mitigate significant losses in yield and quality. Traditional methods of disease identification rely heavily on subjective judgment by farmers or experts, which can lead to inefficiencies and missed opportunities for intervention. Recent advancements in machine learning and deep learning offer promising solutions for automating and improving the accuracy of plant disease detection. This paper presents a comprehensive analysis of state-of-the-art Convolutional Neural Network (CNN) models for classifying diseases in pumpkin plant leaves. Using a publicly available dataset of 2000 highresolution images, we evaluate the performance of several CNN architectures, including ResNet, DenseNet, and EfficientNet, in recognizing five classes: healthy leaves and four common diseases downy mildew, powdery mildew, mosaic disease, and bacterial leaf spot. We fine-tuned these pretrained models and conducted hyperparameter optimization experiments. ResNet-34, DenseNet-121, and EfficientNet-B7 were identified as top-performing models, each excelling in different classes of leaf diseases. Our analysis revealed DenseNet-121 as the optimal model when considering both accuracy and computational complexity achieving an overall accuracy of 86%. This study underscores the potential of CNNs in automating disease diagnosis for pumpkin plants, offering valuable insights that can contribute to enhancing agricultural productivity and minimizing economic losses.
Authors: Kaushik Roy, Akila Dissanayake, Brendan Tidd, Peyman Moghadam
Abstract: Lifelong imitation learning for manipulation tasks poses significant challenges due to distribution shifts that occur in incremental learning steps. Existing methods often focus on unsupervised skill discovery to construct an ever-growing skill library or distillation from multiple policies, which can lead to scalability issues as diverse manipulation tasks are continually introduced and may fail to ensure a consistent latent space throughout the learning process, leading to catastrophic forgetting of previously learned skills. In this paper, we introduce M2Distill, a multi-modal distillation-based method for lifelong imitation learning focusing on preserving consistent latent space across vision, language, and action distributions throughout the learning process. By regulating the shifts in latent representations across different modalities from previous to current steps, and reducing discrepancies in Gaussian Mixture Model (GMM) policies between consecutive learning steps, we ensure that the learned policy retains its ability to perform previously learned tasks while seamlessly integrating new skills. Extensive evaluations on the LIBERO lifelong imitation learning benchmark suites, including LIBERO-OBJECT, LIBERO-GOAL, and LIBERO-SPATIAL, demonstrate that our method consistently outperforms prior state-of-the-art methods across all evaluated metrics.
Authors: Oytun Demirbilek, Tingying Peng, Alaa Bessadok
Abstract: A morphological brain graph depicting a connectional fingerprint is of paramount importance for charting brain dysconnectivity patterns. Such data often has missing observations due to various reasons such as time-consuming and incomplete neuroimage processing pipelines. Thus, predicting a target brain graph from a source graph is crucial for better diagnosing neurological disorders with minimal data acquisition resources. Many brain graph generative models were proposed for promising results, yet they are mostly based on generative adversarial networks (GAN), which could suffer from mode collapse and require large training datasets. Recent developments in diffusion models address these problems by offering essential properties such as a stable training objective and easy scalability. However, applying a diffusion process to graph edges fails to maintain the topological symmetry of the brain connectivity matrices. To meet these challenges, we propose the Graph Residual Noise Learner Network (Grenol-Net), the first graph diffusion model for predicting a target graph from a source graph.
Authors: Giannis Daras, Hyungjin Chung, Chieh-Hsin Lai, Yuki Mitsufuji, Jong Chul Ye, Peyman Milanfar, Alexandros G. Dimakis, Mauricio Delbracio
Abstract: Diffusion models have become increasingly popular for generative modeling due to their ability to generate high-quality samples. This has unlocked exciting new possibilities for solving inverse problems, especially in image restoration and reconstruction, by treating diffusion models as unsupervised priors. This survey provides a comprehensive overview of methods that utilize pre-trained diffusion models to solve inverse problems without requiring further training. We introduce taxonomies to categorize these methods based on both the problems they address and the techniques they employ. We analyze the connections between different approaches, offering insights into their practical implementation and highlighting important considerations. We further discuss specific challenges and potential solutions associated with using latent diffusion models for inverse problems. This work aims to be a valuable resource for those interested in learning about the intersection of diffusion models and inverse problems.
Authors: Sunil Kumar, Ted Sandler, Paulina Varshavskaya
Abstract: Fine-tuning modern computer vision models requires accurately labeled data for which the ground truth may not exist, but a set of multiple labels can be obtained from labelers of variable accuracy. We tie the notion of label quality to confidence in labeler accuracy and show that, when prior estimates of labeler accuracy are available, using a simple naive-Bayes model to estimate the true labels allows us to label more data on a fixed budget without compromising label or fine-tuning quality. We present experiments on a dataset of industrial images that demonstrates that our method, called Ground Truth Extension (GTX), enables fine-tuning ML models using fewer human labels.
Authors: Alan Papalia, Yulun Tian, David M. Rosen, Jonathan P. How, John J. Leonard
Abstract: This paper presents an overview of the Burer-Monteiro method (BM), a technique that has been applied to solve robot perception problems to certifiable optimality in real-time. BM is often used to solve semidefinite programming relaxations, which can be used to perform global optimization for non-convex perception problems. Specifically, BM leverages the low-rank structure of typical semidefinite programs to dramatically reduce the computational cost of performing optimization. This paper discusses BM in certifiable perception, with three main objectives: (i) to consolidate information from the literature into a unified presentation, (ii) to elucidate the role of the linear independence constraint qualification (LICQ), a concept not yet well-covered in certifiable perception literature, and (iii) to share practical considerations that are discussed among practitioners but not thoroughly covered in the literature. Our general aim is to offer a practical primer for applying BM towards certifiable perception.
Authors: Jun Jiang, Raymond Moore, Brenna Novotny, Leo Liu, Zachary Fogarty, Ray Guo, Markovic Svetomir, Chen Wang
Abstract: Histopathological imaging is vital for cancer research and clinical practice, with multiplexed Immunofluorescence (MxIF) and Hematoxylin and Eosin (H&E) providing complementary insights. However, aligning different stains at the cell level remains a challenge due to modality differences. In this paper, we present a novel framework for multimodal image alignment using cell segmentation outcomes. By treating cells as point sets, we apply Coherent Point Drift (CPD) for initial alignment and refine it with Graph Matching (GM). Evaluated on ovarian cancer tissue microarrays (TMAs), our method achieves high alignment accuracy, enabling integration of cell-level features across modalities and generating virtual H&E images from MxIF data for enhanced clinical interpretation.
Authors: Siyeop Yoon, Rui Hu, Yuang Wang, Matthew Tivnan, Young-don Son, Dufan Wu, Xiang Li, Kyungsang Kim, Quanzheng Li
Abstract: PET imaging is a powerful modality offering quantitative assessments of molecular and physiological processes. The necessity for PET denoising arises from the intrinsic high noise levels in PET imaging, which can significantly hinder the accurate interpretation and quantitative analysis of the scans. With advances in deep learning techniques, diffusion model-based PET denoising techniques have shown remarkable performance improvement. However, these models often face limitations when applied to volumetric data. Additionally, many existing diffusion models do not adequately consider the unique characteristics of PET imaging, such as its 3D volumetric nature, leading to the potential loss of anatomic consistency. Our Conditional Score-based Residual Diffusion (CSRD) model addresses these issues by incorporating a refined score function and 3D patch-wise training strategy, optimizing the model for efficient volumetric PET denoising. The CSRD model significantly lowers computational demands and expedites the denoising process. By effectively integrating volumetric data from PET and MRI scans, the CSRD model maintains spatial coherence and anatomical detail. Lastly, we demonstrate that the CSRD model achieves superior denoising performance in both qualitative and quantitative evaluations while maintaining image details and outperforms existing state-of-the-art methods.
Authors: Buse Giledereli, Yifan Hou, Yilei Tu, Mrinmaya Sachan
Abstract: Visual language is a system of communication that conveys information through symbols, shapes, and spatial arrangements. Diagrams are a typical example of a visual language depicting complex concepts and their relationships in the form of an image. The symbolic nature of diagrams presents significant challenges for building models capable of understanding them. Yet, recent studies seem to suggest that Large Vision-Language Models (LVLMs) can even tackle complex reasoning tasks involving diagrams. In this paper, we investigate this phenomenon by developing a comprehensive test suite to evaluate the diagram comprehension capability of LVLMs. Our test suite uses a variety of questions focused on concept entities and their relationships over a set of synthetic as well as real diagrams across several domains to evaluate the recognition and reasoning abilities of models. Our evaluation of three LVLMs (GPT-4V, GPT-4o, and Gemini) shows that while these models can accurately identify and reason about entities, their ability to understand relationships is notably limited. Further testing reveals that the decent performance on diagram understanding largely stems from leveraging their background knowledge as shortcuts to identify and reason about the relational information. Thus, we conclude that LVLMs have a limited capability for genuine diagram understanding, and their impressive performance in diagram reasoning is an illusion emanating from other confounding factors, such as the background knowledge in the models.
Authors: Qi Wu, Zipeng Fu, Xuxin Cheng, Xiaolong Wang, Chelsea Finn
Abstract: Learning-based methods have achieved strong performance for quadrupedal locomotion. However, several challenges prevent quadrupeds from learning helpful indoor skills that require interaction with environments and humans: lack of end-effectors for manipulation, limited semantic understanding using only simulation data, and low traversability and reachability in indoor environments. We present a system for quadrupedal mobile manipulation in indoor environments. It uses a front-mounted gripper for object manipulation, a low-level controller trained in simulation using egocentric depth for agile skills like climbing and whole-body tilting, and pre-trained vision-language models (VLMs) with a third-person fisheye and an egocentric RGB camera for semantic understanding and command generation. We evaluate our system in two unseen environments without any real-world data collection or training. Our system can zero-shot generalize to these environments and complete tasks, like following user's commands to fetch a randomly placed stuff toy after climbing over a queen-sized bed, with a 60% success rate. Project website: https://helpful-doggybot.github.io/
Authors: Weitai Kang, Haifeng Huang, Yuzhang Shang, Mubarak Shah, Yan Yan
Abstract: Recent advancements in 3D Large Language Models (3DLLMs) have highlighted their potential in building general-purpose agents in the 3D real world, yet challenges remain due to the lack of high-quality robust instruction-following data, leading to limited discriminative power and generalization of 3DLLMs. In this paper, we introduce Robin3D, a powerful 3DLLM trained on large-scale instruction-following data generated by our novel data engine, Robust Instruction Generation (RIG) engine. RIG generates two key instruction data: 1) the Adversarial Instruction-following data, which features mixed negative and positive samples to enhance the model's discriminative understanding. 2) the Diverse Instruction-following data, which contains various instruction styles to enhance model's generalization. As a result, we construct 1 million instruction-following data, consisting of 344K Adversarial samples, 508K Diverse samples, and 165K benchmark training set samples. To better handle these complex instructions, Robin3D first incorporates Relation-Augmented Projector to enhance spatial understanding, and then strengthens the object referring and grounding ability through ID-Feature Bonding. Robin3D consistently outperforms previous methods across five widely-used 3D multimodal learning benchmarks, without the need for task-specific fine-tuning. Notably, we achieve a 7.8\% improvement in the grounding task (Multi3DRefer) and a 6.9\% improvement in the captioning task (Scan2Cap).
Authors: Chun-Hsiao Yeh, Jiayun Wang, Andrew D. Graham, Andrea J. Liu, Bo Tan, Yubei Chen, Yi Ma, Meng C. Lin
Abstract: Accurate diagnosis of ocular surface diseases is critical in optometry and ophthalmology, which hinge on integrating clinical data sources (e.g., meibography imaging and clinical metadata). Traditional human assessments lack precision in quantifying clinical observations, while current machine-based methods often treat diagnoses as multi-class classification problems, limiting the diagnoses to a predefined closed-set of curated answers without reasoning the clinical relevance of each variable to the diagnosis. To tackle these challenges, we introduce an innovative multi-modal diagnostic pipeline (MDPipe) by employing large language models (LLMs) for ocular surface disease diagnosis. We first employ a visual translator to interpret meibography images by converting them into quantifiable morphology data, facilitating their integration with clinical metadata and enabling the communication of nuanced medical insight to LLMs. To further advance this communication, we introduce a LLM-based summarizer to contextualize the insight from the combined morphology and clinical metadata, and generate clinical report summaries. Finally, we refine the LLMs' reasoning ability with domain-specific insight from real-life clinician diagnoses. Our evaluation across diverse ocular surface disease diagnosis benchmarks demonstrates that MDPipe outperforms existing standards, including GPT-4, and provides clinically sound rationales for diagnoses.
Authors: Xueming Fu, Yingtai Li, Fenghe Tang, Jun Li, Mingyue Zhao, Gao-Jun Teng, S. Kevin Zhou
Abstract: Reconstructing 3D coronary arteries is important for coronary artery disease diagnosis, treatment planning and operation navigation. Traditional reconstruction techniques often require many projections, while reconstruction from sparse-view X-ray projections is a potential way of reducing radiation dose. However, the extreme sparsity of coronary arteries in a 3D volume and ultra-limited number of projections pose significant challenges for efficient and accurate 3D reconstruction. To this end, we propose 3DGR-CAR, a 3D Gaussian Representation for Coronary Artery Reconstruction from ultra-sparse X-ray projections. We leverage 3D Gaussian representation to avoid the inefficiency caused by the extreme sparsity of coronary artery data and propose a Gaussian center predictor to overcome the noisy Gaussian initialization from ultra-sparse view projections. The proposed scheme enables fast and accurate 3D coronary artery reconstruction with only 2 views. Experimental results on two datasets indicate that the proposed approach significantly outperforms other methods in terms of voxel accuracy and visual quality of coronary arteries. The code will be available in https://github.com/windrise/3DGR-CAR.
Authors: Jonghun Kim, Mansu Kim, Hyunjin Park
Abstract: The scarcity of annotated medical images is a major bottleneck in developing learning models for medical image analysis. Hence, recent studies have focused on pretrained models with fewer annotation requirements that can be fine-tuned for various downstream tasks. However, existing approaches are mainly 3D adaptions of 2D approaches ill-suited for 3D medical imaging data. Motivated by this gap, we propose novel domain-aware multi-task learning tasks to pretrain a 3D Swin Transformer for brain magnetic resonance imaging (MRI). Our method considers the domain knowledge in brain MRI by incorporating brain anatomy and morphology as well as standard pretext tasks adapted for 3D imaging in a contrastive learning setting. We pretrain our model using large-scale brain MRI data of 13,687 samples spanning several large-scale databases. Our method outperforms existing supervised and self-supervised methods in three downstream tasks of Alzheimer's disease classification, Parkinson's disease classification, and age prediction tasks. The ablation study of the proposed pretext tasks shows the effectiveness of our pretext tasks.
Authors: Guy Ohayon, Tomer Michaeli, Michael Elad
Abstract: Photo-realistic image restoration algorithms are typically evaluated by distortion measures (e.g., PSNR, SSIM) and by perceptual quality measures (e.g., FID, NIQE), where the desire is to attain the lowest possible distortion without compromising on perceptual quality. To achieve this goal, current methods typically attempt to sample from the posterior distribution, or to optimize a weighted sum of a distortion loss (e.g., MSE) and a perceptual quality loss (e.g., GAN). Unlike previous works, this paper is concerned specifically with the optimal estimator that minimizes the MSE under a constraint of perfect perceptual index, namely where the distribution of the reconstructed images is equal to that of the ground-truth ones. A recent theoretical result shows that such an estimator can be constructed by optimally transporting the posterior mean prediction (MMSE estimate) to the distribution of the ground-truth images. Inspired by this result, we introduce Posterior-Mean Rectified Flow (PMRF), a simple yet highly effective algorithm that approximates this optimal estimator. In particular, PMRF first predicts the posterior mean, and then transports the result to a high-quality image using a rectified flow model that approximates the desired optimal transport map. We investigate the theoretical utility of PMRF and demonstrate that it consistently outperforms previous methods on a variety of image restoration tasks.
Authors: Miyu Goko, Motonari Kambara, Daichi Saito, Seitaro Otsuki, Komei Sugiura
Abstract: In this study, we consider the problem of predicting task success for open-vocabulary manipulation by a manipulator, based on instruction sentences and egocentric images before and after manipulation. Conventional approaches, including multimodal large language models (MLLMs), often fail to appropriately understand detailed characteristics of objects and/or subtle changes in the position of objects. We propose Contrastive $\lambda$-Repformer, which predicts task success for table-top manipulation tasks by aligning images with instruction sentences. Our method integrates the following three key types of features into a multi-level aligned representation: features that preserve local image information; features aligned with natural language; and features structured through natural language. This allows the model to focus on important changes by looking at the differences in the representation between two images. We evaluate Contrastive $\lambda$-Repformer on a dataset based on a large-scale standard dataset, the RT-1 dataset, and on a physical robot platform. The results show that our approach outperformed existing approaches including MLLMs. Our best model achieved an improvement of 8.66 points in accuracy compared to the representative MLLM-based model.
Authors: Krzysztof Zieli\'nski, Bruce Blumberg, Mikkel Baun Kj{\ae}rgaard
Abstract: Capturing real-world 3D spaces as point clouds is efficient and descriptive, but it comes with sensor errors and lacks object parametrization. These limitations render point clouds unsuitable for various real-world applications, such as robot programming, without extensive post-processing (e.g., outlier removal, semantic segmentation). On the other hand, CAD modeling provides high-quality, parametric representations of 3D space with embedded semantic data, but requires manual component creation that is time-consuming and costly. To address these challenges, we propose a novel solution that combines the strengths of both approaches. Our method for 3D workcell sketching from point clouds allows users to refine raw point clouds using an Augmented Reality (AR) interface that leverages their knowledge and the real-world 3D environment. By utilizing a toolbox and an AR-enabled pointing device, users can enhance point cloud accuracy based on the device's position in 3D space. We validate our approach by comparing it with ground truth models, demonstrating that it achieves a mean error within 1cm - significant improvement over standard LiDAR scanner apps.
Authors: Yuchi Ishikawa, Tatsuya Komatsu, Yoshimitsu Aoki
Abstract: In this paper, we propose to pre-train audio encoders using synthetic patterns instead of real audio data. Our proposed framework consists of two key elements. The first one is Masked Autoencoder (MAE), a self-supervised learning framework that learns from reconstructing data from randomly masked counterparts. MAEs tend to focus on low-level information such as visual patterns and regularities within data. Therefore, it is unimportant what is portrayed in the input, whether it be images, audio mel-spectrograms, or even synthetic patterns. This leads to the second key element, which is synthetic data. Synthetic data, unlike real audio, is free from privacy and licensing infringement issues. By combining MAEs and synthetic patterns, our framework enables the model to learn generalized feature representations without real data, while addressing the issues related to real audio. To evaluate the efficacy of our framework, we conduct extensive experiments across a total of 13 audio tasks and 17 synthetic datasets. The experiments provide insights into which types of synthetic patterns are effective for audio. Our results demonstrate that our framework achieves performance comparable to models pre-trained on AudioSet-2M and partially outperforms image-based pre-training methods.
Authors: Patrick Kramer, Alexander Steinhardt, Barbara Pedretscher
Abstract: This paper investigates the enhancement of spatial resolution in Sentinel-2 bands that contain spectral information using advanced super-resolution techniques by a factor of 2. State-of-the-art CNN models are compared with enhanced GAN approaches in terms of quality and feasibility. Therefore, a representative dataset comprising Sentinel-2 low-resolution images and corresponding high-resolution aerial orthophotos is required. Literature study reveals no feasible dataset for the land type of interest (forests), for which reason an adequate dataset had to be generated in addition, accounting for accurate alignment and image source optimization. The results reveal that while CNN-based approaches produce satisfactory outcomes, they tend to yield blurry images. In contrast, GAN-based models not only provide clear and detailed images, but also demonstrate superior performance in terms of quantitative assessment, underlying the potential of the framework beyond the specific land type investigated.
Authors: Taewook Park, Seunghwan Kim, Hyondong Oh
Abstract: Reliable perception of targets is crucial for the stable operation of autonomous robots. A widely preferred method is keypoint identification in an image, as it allows direct mapping from raw images to 2D coordinates, facilitating integration with other algorithms like localization and path planning. In this study, we closely examine the design and identification of keypoint patches in cluttered environments, where factors such as blur and shadows can hinder detection. We propose four simple yet distinct designs that consider various scale, rotation and camera projection using a limited number of pixels. Additionally, we customize the Superpoint network to ensure robust detection under various types of image degradation. The effectiveness of our approach is demonstrated through real-world video tests, highlighting potential for vision-based autonomous systems.
Authors: Hamed Behzadi-Khormouji, Jos\'e Oramas
Abstract: Model Interpretation aims at the extraction of insights from the internals of a trained model. A common approach to address this task is the characterization of relevant features internally encoded in the model that are critical for its proper operation. Despite recent progress of these methods, they come with the weakness of being computationally expensive due to the dense evaluation of datasets that they require. As a consequence, research on the design of these methods have focused on smaller data subsets which may led to reduced insights. To address these computational costs, we propose a coreset-based interpretation framework that utilizes coreset selection methods to sample a representative subset of the large dataset for the interpretation task. Towards this goal, we propose a similarity-based evaluation protocol to assess the robustness of model interpretation methods towards the amount data they take as input. Experiments considering several interpretation methods, DNN models, and coreset selection methods show the effectiveness of the proposed framework.
Authors: Krishna Chaitanya, Pablo F. Damasceno, Shreyas Fadnavis, Pooya Mobadersany, Chaitanya Parmar, Emily Scherer, Natalia Zemlianskaia, Lindsey Surace, Louis R. Ghanem, Oana Gabriela Cula, Tommaso Mansi, Kristopher Standish
Abstract: Accurate assessment of disease severity from endoscopy videos in ulcerative colitis (UC) is crucial for evaluating drug efficacy in clinical trials. Severity is often measured by the Mayo Endoscopic Subscore (MES) and Ulcerative Colitis Endoscopic Index of Severity (UCEIS) score. However, expert MES/UCEIS annotation is time-consuming and susceptible to inter-rater variability, factors addressable by automation. Automation attempts with frame-level labels face challenges in fully-supervised solutions due to the prevalence of video-level labels in clinical trials. CNN-based weakly-supervised models (WSL) with end-to-end (e2e) training lack generalization to new disease scores and ignore spatio-temporal information crucial for accurate scoring. To address these limitations, we propose "Arges", a deep learning framework that utilizes a transformer with positional encoding to incorporate spatio-temporal information from frame features to estimate disease severity scores in endoscopy video. Extracted features are derived from a foundation model (ArgesFM), pre-trained on a large diverse dataset from multiple clinical trials (61M frames, 3927 videos). We evaluate four UC disease severity scores, including MES and three UCEIS component scores. Test set evaluation indicates significant improvements, with F1 scores increasing by 4.1% for MES and 18.8%, 6.6%, 3.8% for the three UCEIS component scores compared to state-of-the-art methods. Prospective validation on previously unseen clinical trial data further demonstrates the model's successful generalization.
Authors: Xingdou Fu, Lin Miao, Yasuhiro Ohnishi, Yuki Hasegawa, Masaki Suwa
Abstract: Bin picking systems in factory automation usually face robustness issues caused by sparse and noisy 3D data of metallic objects. Utilizing multiple views, especially with a one-shot 3D sensor and "sensor on hand" configuration is getting more popularity due to its effectiveness, flexibility, and low cost. While moving the 3D sensor to acquire multiple views for 3D fusion, joint optimization, or active vision suffers from low-speed issues. That is because sensing is taken as a decoupled module from motion tasks and is not intentionally designed for a bin picking system. To address the problems, we designed a bin picking system, which tightly couples a multi-view, active vision scheme with motion tasks in a "sensor on hand" configuration. It not only speeds up the system by parallelizing the high-speed sensing scheme to the robot place action but also decides the next sensing path to maintain the continuity of the whole picking process. Unlike others focusing only on sensing evaluation, we also evaluated our design by picking experiments on 5 different types of objects without human intervention. Our experiments show the whole sensing scheme can be finished within 1.682 seconds (maximum) on CPU and the average picking complete rate is over 97.75%. Due to the parallelization with robot motion, the sensing scheme accounts for only 0.635 seconds in takt time on average.
Authors: Jiapeng Wang, Chengyu Wang, Kunzhe Huang, Jun Huang, Lianwen Jin
Abstract: Contrastive Language-Image Pre-training (CLIP) has been widely studied and applied in numerous applications. However, the emphasis on brief summary texts during pre-training prevents CLIP from understanding long descriptions. This issue is particularly acute regarding videos given that videos often contain abundant detailed contents. In this paper, we propose the VideoCLIP-XL (eXtra Length) model, which aims to unleash the long-description understanding capability of video CLIP models. Firstly, we establish an automatic data collection system and gather a large-scale VILD pre-training dataset with VIdeo and Long-Description pairs. Then, we propose Text-similarity-guided Primary Component Matching (TPCM) to better learn the distribution of feature space while expanding the long description capability. We also introduce two new tasks namely Detail-aware Description Ranking (DDR) and Hallucination-aware Description Ranking (HDR) for further understanding improvement. Finally, we construct a Long Video Description Ranking (LVDR) benchmark for evaluating the long-description capability more comprehensively. Extensive experimental results on widely-used text-video retrieval benchmarks with both short and long descriptions and our LVDR benchmark can fully demonstrate the effectiveness of our method.
Authors: Paul Weiser, Georg Langs, Stanislav Motyka, Wolfgang Bogner, S\'ebastien Courvoisier, Malte Hoffmann, Antoine Klauser, Ovidiu C. Andronesi
Abstract: Purpose. Proton Magnetic Resonance Spectroscopic Imaging (1H-MRSI) provides non-invasive spectral-spatial mapping of metabolism. However, long-standing problems in whole-brain 1H-MRSI are spectral overlap of metabolite peaks with large lipid signal from scalp, and overwhelming water signal that distorts spectra. Fast and effective methods are needed for high-resolution 1H-MRSI to accurately remove lipid and water signals while preserving the metabolite signal. The potential of supervised neural networks for this task remains unexplored, despite their success for other MRSI processing. Methods. We introduce a deep-learning method based on a modified Y-NET network for water and lipid removal in whole-brain 1H-MRSI. The WALINET (WAter and LIpid neural NETwork) was compared to conventional methods such as the state-of-the-art lipid L2 regularization and Hankel-Lanczos singular value decomposition (HLSVD) water suppression. Methods were evaluated on simulated and in-vivo whole-brain MRSI using NMRSE, SNR, CRLB, and FWHM metrics. Results. WALINET is significantly faster and needs 8s for high-resolution whole-brain MRSI, compared to 42 minutes for conventional HLSVD+L2. Quantitative analysis shows WALINET has better performance than HLSVD+L2: 1) more lipid removal with 41% lower NRMSE, 2) better metabolite signal preservation with 71% lower NRMSE in simulated data, 155% higher SNR and 50% lower CRLB in in-vivo data. Metabolic maps obtained by WALINET in healthy subjects and patients show better gray/white-matter contrast with more visible structural details. Conclusions. WALINET has superior performance for nuisance signal removal and metabolite quantification on whole-brain 1H-MRSI compared to conventional state-of-the-art techniques. This represents a new application of deep-learning for MRSI processing, with potential for automated high-throughput workflow.
Authors: Rui Tang, Shirong Guo, Yuhang Qiu, Honghui Chen, Lujin Huang, Ming Yong, Linfu Zhou, Liquan Guo
Abstract: Robots-based smart pharmacies are essential for modern healthcare systems, enabling efficient drug delivery. However, a critical challenge exists in the robotic handling of drugs with varying shapes and overlapping positions, which previous studies have not adequately addressed. To enhance the robotic arm's ability to grasp chaotic, overlapping, and variously shaped drugs, this paper proposed a novel framework combining a multi-stage grasping network with an adaptive robotics mechanism. The framework first preprocessed images using an improved Super-Resolution Convolutional Neural Network (SRCNN) algorithm, and then employed the proposed YOLOv5+E-A-SPPFCSPC+BIFPNC (YOLO-EASB) instance segmentation algorithm for precise drug segmentation. The most suitable drugs for grasping can be determined by assessing the completeness of the segmentation masks. Then, these segmented drugs were processed by our improved Adaptive Feature Fusion and Grasp-Aware Network (IAFFGA-Net) with the optimized loss function, which ensures accurate picking actions even in complex environments. To control the robot grasping, a time-optimal robotic arm trajectory planning algorithm that combines an improved ant colony algorithm with 3-5-3 interpolation was developed, further improving efficiency while ensuring smooth trajectories. Finally, this system was implemented and validated within an adaptive collaborative robot setup, which dynamically adjusts to different production environments and task requirements. Experimental results demonstrate the superiority of our multi-stage grasping network in optimizing smart pharmacy operations, while also showcasing its remarkable adaptability and effectiveness in practical applications.
Authors: Mohammad Abuzar Hashemi, Zhanghexuan Li, Mihir Chauhan, Yan Shen, Abhishek Satbhai, Mir Basheer Ali, Mingchen Gao, Sargur Srihari
Abstract: Pre-training visual and textual representations from large-scale image-text pairs is becoming a standard approach for many downstream vision-language tasks. The transformer-based models learn inter and intra-modal attention through a list of self-supervised learning tasks. This paper proposes LAViTeR, a novel architecture for visual and textual representation learning. The main module, Visual Textual Alignment (VTA) will be assisted by two auxiliary tasks, GAN-based image synthesis and Image Captioning. We also propose a new evaluation metric measuring the similarity between the learnt visual and textual embedding. The experimental results on two public datasets, CUB and MS-COCO, demonstrate superior visual and textual representation alignment in the joint feature embedding space
Authors: Xiao Han, Yiming Ren, Peishan Cong, Yujing Sun, Jingya Wang, Lan Xu, Yuexin Ma
Abstract: Human gait recognition is crucial in multimedia, enabling identification through walking patterns without direct interaction, enhancing the integration across various media forms in real-world applications like smart homes, healthcare and non-intrusive security. LiDAR's ability to capture depth makes it pivotal for robotic perception and holds promise for real-world gait recognition. In this paper, based on a single LiDAR, we present the Hierarchical Multi-representation Feature Interaction Network (HMRNet) for robust gait recognition. Prevailing LiDAR-based gait datasets primarily derive from controlled settings with predefined trajectory, remaining a gap with real-world scenarios. To facilitate LiDAR-based gait recognition research, we introduce FreeGait, a comprehensive gait dataset from large-scale, unconstrained settings, enriched with multi-modal and varied 2D/3D data. Notably, our approach achieves state-of-the-art performance on prior dataset (SUSTech1K) and on FreeGait.
Authors: Minjoon Jung, Youwon Jang, Seongho Choi, Joochan Kim, Jin-Hwa Kim, Byoung-Tak Zhang
Abstract: Video moment retrieval (VMR) identifies a specific moment in an untrimmed video for a given natural language query. This task is prone to suffer the weak alignment problem innate in video datasets. Due to the ambiguity, a query does not fully cover the relevant details of the corresponding moment, or the moment may contain misaligned and irrelevant frames, potentially limiting further performance gains. To tackle this problem, we propose a background-aware moment detection transformer (BM-DETR). Our model adopts a contrastive approach, carefully utilizing the negative queries matched to other moments in the video. Specifically, our model learns to predict the target moment from the joint probability of each frame given the positive query and the complement of negative queries. This leads to effective use of the surrounding background, improving moment sensitivity and enhancing overall alignments in videos. Extensive experiments on four benchmarks demonstrate the effectiveness of our approach. Our code is available at: \url{https://github.com/minjoong507/BM-DETR}
Authors: Nikita Karaev, Ignacio Rocco, Benjamin Graham, Natalia Neverova, Andrea Vedaldi, Christian Rupprecht
Abstract: We introduce CoTracker, a transformer-based model that tracks a large number of 2D points in long video sequences. Differently from most existing approaches that track points independently, CoTracker tracks them jointly, accounting for their dependencies. We show that joint tracking significantly improves tracking accuracy and robustness, and allows CoTracker to track occluded points and points outside of the camera view. We also introduce several innovations for this class of trackers, including using token proxies that significantly improve memory efficiency and allow CoTracker to track 70k points jointly and simultaneously at inference on a single GPU. CoTracker is an online algorithm that operates causally on short windows. However, it is trained utilizing unrolled windows as a recurrent network, maintaining tracks for long periods of time even when points are occluded or leave the field of view. Quantitatively, CoTracker substantially outperforms prior trackers on standard point-tracking benchmarks.
Authors: Mengyuan Liu, Hong Liu, Tianyu Guo
Abstract: Considering the instance-level discriminative ability, contrastive learning methods, including MoCo and SimCLR, have been adapted from the original image representation learning task to solve the self-supervised skeleton-based action recognition task. These methods usually use multiple data streams (i.e., joint, motion, and bone) for ensemble learning, meanwhile, how to construct a discriminative feature space within a single stream and effectively aggregate the information from multiple streams remains an open problem. To this end, this paper first applies a new contrastive learning method called BYOL to learn from skeleton data, and then formulate SkeletonBYOL as a simple yet effective baseline for self-supervised skeleton-based action recognition. Inspired by SkeletonBYOL, this paper further presents a Cross-Model and Cross-Stream (CMCS) framework. This framework combines Cross-Model Adversarial Learning (CMAL) and Cross-Stream Collaborative Learning (CSCL). Specifically, CMAL learns single-stream representation by cross-model adversarial loss to obtain more discriminative features. To aggregate and interact with multi-stream information, CSCL is designed by generating similarity pseudo label of ensemble learning as supervision and guiding feature generation for individual streams. Extensive experiments on three datasets verify the complementary properties between CMAL and CSCL and also verify that the proposed method can achieve better results than state-of-the-art methods using various evaluation protocols.
Authors: Akshaj Gaur, Cheng Liu, Xiaomin Lin, Nare Karapetyan, Yiannis Aloimonos
Abstract: With a number of marine populations in rapid decline, collecting and analyzing data about marine populations has become increasingly important to develop effective conservation policies for a wide range of marine animals, including whales. Modern computer vision algorithms allow us to detect whales in images in a wide range of domains, further speeding up and enhancing the monitoring process. However, these algorithms heavily rely on large training datasets, which are challenging and time-consuming to collect particularly in marine or aquatic environments. Recent advances in AI however have made it possible to synthetically create datasets for training machine learning algorithms, thus enabling new solutions that were not possible before. In this work, we present a solution - SeaDroneSim2 benchmark suite, which addresses this challenge by generating aerial, and satellite synthetic image datasets to improve the detection of whales and reduce the effort required for training data collection. We show that we can achieve a 15% performance boost on whale detection compared to using the real data alone for training, by augmenting a 10% real data. We open source both the code of the simulation platform SeaDroneSim2 and the dataset generated through it.
Authors: Changming Xiao, Qi Yang, Feng Zhou, Changshui Zhang
Abstract: Diffusion models have revolted the field of text-to-image generation recently. The unique way of fusing text and image information contributes to their remarkable capability of generating highly text-related images. From another perspective, these generative models imply clues about the precise correlation between words and pixels. In this work, a simple but effective method is proposed to utilize the attention mechanism in the denoising network of text-to-image diffusion models. Without re-training nor inference-time optimization, the semantic grounding of phrases can be attained directly. We evaluate our method on Pascal VOC 2012 and Microsoft COCO 2014 under weakly-supervised semantic segmentation setting and our method achieves superior performance to prior methods. In addition, the acquired word-pixel correlation is found to be generalizable for the learned text embedding of customized generation methods, requiring only a few modifications. To validate our discovery, we introduce a new practical task called "personalized referring image segmentation" with a new dataset. Experiments in various situations demonstrate the advantages of our method compared to strong baselines on this task. In summary, our work reveals a novel way to extract the rich multi-modal knowledge hidden in diffusion models for segmentation.
Authors: Ahmed Zidane, Ghada Soliman
Abstract: This paper proposes an approach of Ladder Bottom-up Convolutional Bidirectional Variational Autoencoder (LCBVAE) architecture for the encoder and decoder, which is trained on the image translation of the dotted Arabic expiration dates by reconstructing the Arabic dotted expiration dates into filled-in expiration dates. We employed a customized and adapted version of Convolutional Recurrent Neural Network CRNN model to meet our specific requirements and enhance its performance in our context, and then trained the custom CRNN model with the filled-in images from the year of 2019 to 2027 to extract the expiration dates and assess the model performance of LCBVAE on the expiration date recognition. The pipeline of (LCBVAE+CRNN) can be then integrated into an automated sorting systems for extracting the expiry dates and sorting the products accordingly during the manufacture stage. Additionally, it can overcome the manual entry of expiration dates that can be time-consuming and inefficient at the merchants. Due to the lack of the availability of the dotted Arabic expiration date images, we created an Arabic dot-matrix True Type Font (TTF) for the generation of the synthetic images. We trained the model with unrealistic synthetic dates of 60,000 images and performed the testing on a realistic synthetic date of 3000 images from the year of 2019 to 2027, represented as yyyy/mm/dd. In our study, we demonstrated the significance of latent bottleneck layer with improving the generalization when the size is increased up to 1024 in downstream transfer learning tasks as for image translation. The proposed approach achieved an accuracy of 97% on the image translation with using the LCBVAE architecture that can be generalized for any downstream learning tasks as for image translation and reconstruction.
Authors: Onur N. Tepencelik, Wenchuan Wei, Pamela C. Cosman, Sujit Dey
Abstract: We propose a system that estimates people's body and head orientations using low-resolution point cloud data from two LiDAR sensors. Our models make accurate estimations in real-world conversation settings where subjects move naturally with varying head and body poses, while seated around a table. The body orientation estimation model uses ellipse fitting while the head orientation estimation model combines geometric feature extraction with an ensemble of neural network regressors. Our models achieve a mean absolute estimation error of 5.2 degrees for body orientation and 13.7 degrees for head orientation. Compared to other body/head orientation estimation systems that use RGB cameras, our proposed system uses LiDAR sensors to preserve user privacy, while achieving comparable accuracy. Unlike other body/head orientation estimation systems, our sensors do not require a specified close-range placement in front of the subject, enabling estimation from a surveillance viewpoint which produces low-resolution data. This work is the first to attempt head orientation estimation using point clouds in a low-resolution surveillance setting. We compare our model to two state-of-the-art head orientation estimation models that are designed for high-resolution point clouds, which yield higher estimation errors on our low-resolution dataset. We also present an application of head orientation estimation by quantifying behavioral differences between neurotypical and autistic individuals in triadic (three-way) conversations. Significance tests show that autistic individuals display significantly different behavior compared to neurotypical individuals in distributing attention between conversational parties, suggesting that the approach could be a component of a behavioral analysis or coaching system.
Authors: Bin Lin, Yang Ye, Bin Zhu, Jiaxi Cui, Munan Ning, Peng Jin, Li Yuan
Abstract: The Large Vision-Language Model (LVLM) has enhanced the performance of various downstream tasks in visual-language understanding. Most existing approaches encode images and videos into separate feature spaces, which are then fed as inputs to large language models. However, due to the lack of unified tokenization for images and videos, namely misalignment before projection, it becomes challenging for a Large Language Model (LLM) to learn multi-modal interactions from several poor projection layers. In this work, we unify visual representation into the language feature space to advance the foundational LLM towards a unified LVLM. As a result, we establish a simple but robust LVLM baseline, Video-LLaVA, which learns from a mixed dataset of images and videos, mutually enhancing each other. Video-LLaVA achieves superior performances on a broad range of 9 image benchmarks across 5 image question-answering datasets and 4 image benchmark toolkits. Additionally, our Video-LLaVA also outperforms Video-ChatGPT by 5.8%, 9.9%, 18.6%, and 10.1% on MSRVTT, MSVD, TGIF, and ActivityNet, respectively. Notably, extensive experiments demonstrate that Video-LLaVA mutually benefits images and videos within a unified visual representation, outperforming models designed specifically for images or videos. We aim for this work to provide modest insights into the multi-modal inputs for the LLM. Code address: \href{https://github.com/PKU-YuanGroup/Video-LLaVA}
Authors: Siyuan Huang, Ram Prabhakar, Yuxiang Guo, Rama Chellappa, Cheng Peng
Abstract: Person Re-identification is a research area with significant real world applications. Despite recent progress, existing methods face challenges in robust re-identification in the wild, e.g., by focusing only on a particular modality and on unreliable patterns such as clothing. A generalized method is highly desired, but remains elusive to achieve due to issues such as the trade-off between spatial and temporal resolution and imperfect feature extraction. We propose VILLS (Video-Image Learning to Learn Semantics), a self-supervised method that jointly learns spatial and temporal features from images and videos. VILLS first designs a local semantic extraction module that adaptively extracts semantically consistent and robust spatial features. Then, VILLS designs a unified feature learning and adaptation module to represent image and video modalities in a consistent feature space. By Leveraging self-supervised, large-scale pre-training, VILLS establishes a new State-of-The-Art that significantly outperforms existing image and video-based methods.
Authors: Joseph Paul Cohen, Louis Blankemeier, Akshay Chaudhari
Abstract: Models driven by spurious correlations often yield poor generalization performance. We propose the counterfactual (CF) alignment method to detect and quantify spurious correlations of black box classifiers. Our methodology is based on counterfactual images generated with respect to one classifier being input into other classifiers to see if they also induce changes in the outputs of these classifiers. The relationship between these responses can be quantified and used to identify specific instances where a spurious correlation exists. This is validated by observing intuitive trends in a face-attribute face-attribute and waterbird classifiers, as well as by fabricating spurious correlations and detecting their presence, both visually and quantitatively. Furthermore, utilizing the CF alignment method, we demonstrate that we can evaluate robust optimization methods (GroupDRO, JTT, and FLAC) by detecting a reduction in spurious correlations.
Authors: Genki Kinoshita, Ko Nishino
Abstract: In this paper, we introduce a novel training method for making any monocular depth network learn absolute scale and estimate metric road-scene depth just from regular training data, i.e., driving videos. We refer to this training framework as FUMET. The key idea is to leverage cars found on the road as sources of scale supervision and to incorporate them in network training robustly. FUMET detects and estimates the sizes of cars in a frame and aggregates scale information extracted from them into an estimate of the camera height whose consistency across the entire video sequence is enforced as scale supervision. This realizes robust unsupervised training of any, otherwise scale-oblivious, monocular depth network so that they become not only scale-aware but also metric-accurate without the need for auxiliary sensors and extra supervision. Extensive experiments on the KITTI and the Cityscapes datasets show the effectiveness of FUMET, which achieves state-of-the-art accuracy. We also show that FUMET enables training on mixed datasets of different camera heights, which leads to larger-scale training and better generalization. Metric depth reconstruction is essential in any road-scene visual modeling, and FUMET democratizes its deployment by establishing the means to convert any model into a metric depth estimator.
Authors: Xijun Wang, Junbang Liang, Chun-Kai Wang, Kenan Deng, Yu Lou, Ming Lin, Shan Yang
Abstract: In this work, we propose an efficient Video-Language Alignment (ViLA) network. Our ViLA model addresses both efficient frame sampling and effective cross-modal alignment in a unified way. In our ViLA network, we design a new learnable text-guided Frame-Prompter together with a new cross-modal distillation (QFormer-Distiller) module. Pre-trained large image-language models have shown promising results on problems such as visual question answering (VQA). However, how to efficiently and effectively sample video frames when adapting pre-trained large image-language model to video-language alignment is still the major challenge. Compared with prior work, our ViLA model demonstrates the capability of selecting key frames with critical contents, thus improving the video-language alignment accuracy while reducing the inference latency +3.3% on NExT-QA Temporal with 3.0X speed up). Overall, our ViLA network outperforms the state-of-the-art methods on the video question-answering benchmarks: +4.6% on STAR Interaction, +2.2% on STAR average with 3.0X speed up, ours 2-frames out-perform SeViLA 4-frames on the VLEP dataset with 4.2X speed-up. The code will be available at https://github.com/xijun-cs/ViLA.
Authors: Weilong Chai, DanDan Zheng, Jiajiong Cao, Zhiquan Chen, Changbao Wang, Chenguang Ma
Abstract: Text-to-image diffusion models (SD) exhibit significant advancements while requiring extensive computational resources. Existing acceleration methods usually require extensive training and are not universally applicable. LCM-LoRA, trainable once for diverse models, offers universality but rarely considers ensuring the consistency of generated content before and after acceleration. This paper proposes SpeedUpNet (SUN), an innovative acceleration module, to address the challenges of universality and consistency. Exploiting the role of cross-attention layers in U-Net for SD models, we introduce an adapter specifically designed for these layers, quantifying the offset in image generation caused by negative prompts relative to positive prompts. This learned offset demonstrates stability across a range of models, enhancing SUN's universality. To improve output consistency, we propose a Multi-Step Consistency (MSC) loss, which stabilizes the offset and ensures fidelity in accelerated content. Experiments on SD v1.5 show that SUN leads to an overall speedup of more than 10 times compared to the baseline 25-step DPM-solver++, and offers two extra advantages: (1) training-free integration into various fine-tuned Stable-Diffusion models and (2) state-of-the-art FIDs of the generated data set before and after acceleration guided by random combinations of positive and negative prompts. Code is available: https://williechai.github.io/speedup-plugin-for-stable-diffusions.github.io.
URLs: https://williechai.github.io/speedup-plugin-for-stable-diffusions.github.io.
Authors: Licai Sun, Zheng Lian, Kexin Wang, Yu He, Mingyu Xu, Haiyang Sun, Bin Liu, Jianhua Tao
Abstract: Video-based facial affect analysis has recently attracted increasing attention owing to its critical role in human-computer interaction. Previous studies mainly focus on developing various deep learning architectures and training them in a fully supervised manner. Although significant progress has been achieved by these supervised methods, the longstanding lack of large-scale high-quality labeled data severely hinders their further improvements. Motivated by the recent success of self-supervised learning in computer vision, this paper introduces a self-supervised approach, termed Self-supervised Video Facial Affect Perceiver (SVFAP), to address the dilemma faced by supervised methods. Specifically, SVFAP leverages masked facial video autoencoding to perform self-supervised pre-training on massive unlabeled facial videos. Considering that large spatiotemporal redundancy exists in facial videos, we propose a novel temporal pyramid and spatial bottleneck Transformer as the encoder of SVFAP, which not only largely reduces computational costs but also achieves excellent performance. To verify the effectiveness of our method, we conduct experiments on nine datasets spanning three downstream tasks, including dynamic facial expression recognition, dimensional emotion recognition, and personality recognition. Comprehensive results demonstrate that SVFAP can learn powerful affect-related representations via large-scale self-supervised pre-training and it significantly outperforms previous state-of-the-art methods on all datasets. Code is available at https://github.com/sunlicai/SVFAP.
Authors: Qinglong Huang, Haoran Li, Yong Liao, Yanbin Hao, Pengyuan Zhou
Abstract: Neural Radiance Field (NeRF) has been proposed as an innovative advancement in 3D reconstruction techniques. However, little research has been conducted on the issues of information confidentiality and security to NeRF, such as steganography. Existing NeRF steganography solutions have shortcomings in low steganography quality, model weight damage, and limited amount of steganographic information. This paper proposes Noise-NeRF, a novel NeRF steganography method employing Adaptive Pixel Selection strategy and Pixel Perturbation strategy to improve the quality and efficiency of steganography via trainable noise. Extensive experiments validate the state-of-the-art performances of Noise-NeRF on both steganography quality and rendering quality, as well as effectiveness in super-resolution image steganography.
Authors: Qingyuan Yang, Guanzhou Chen, Xiaoliang Tan, Tong Wang, Jiaqi Wang, Xiaodong Zhang
Abstract: Stereo matching and semantic segmentation are significant tasks in binocular satellite 3D reconstruction. However, previous studies primarily view these as independent parallel tasks, lacking an integrated multitask learning framework. This work introduces a solution, the Single-branch Semantic Stereo Network (S3Net), which innovatively combines semantic segmentation and stereo matching using Self-Fuse and Mutual-Fuse modules. Unlike preceding methods that utilize semantic or disparity information independently, our method dentifies and leverages the intrinsic link between these two tasks, leading to a more accurate understanding of semantic information and disparity estimation. Comparative testing on the US3D dataset proves the effectiveness of our S3Net. Our model improves the mIoU in semantic segmentation from 61.38 to 67.39, and reduces the D1-Error and average endpoint error (EPE) in disparity estimation from 10.051 to 9.579 and 1.439 to 1.403 respectively, surpassing existing competitive methods. Our codes are available at:https://github.com/CVEO/S3Net.
Authors: Emilio Morales-Juarez, Gibran Fuentes-Pineda
Abstract: Although the capacity of deep generative models for image generation, such as Diffusion Models (DMs) and Generative Adversarial Networks (GANs), has dramatically improved in recent years, much of their success can be attributed to computationally expensive architectures. This has limited their adoption and use to research laboratories and companies with large resources, while significantly raising the carbon footprint for training, fine-tuning, and inference. In this work, we present LadaGAN, an efficient generative adversarial network that is built upon a novel Transformer block named Ladaformer. The main component of this block is a linear additive-attention mechanism that computes a single attention vector per head instead of the quadratic dot-product attention. We employ Ladaformer in both the generator and discriminator, which reduces the computational complexity and overcomes the training instabilities often associated with Transformer GANs. LadaGAN consistently outperforms existing convolutional and Transformer GANs on benchmark datasets at different resolutions while being significantly more efficient. Moreover, LadaGAN shows competitive performance compared to state-of-the-art multi-step generative models (e.g. DMs) using orders of magnitude less computational resources.
Authors: Jianzong Wu, Xiangtai Li, Chenyang Si, Shangchen Zhou, Jingkang Yang, Jiangning Zhang, Yining Li, Kai Chen, Yunhai Tong, Ziwei Liu, Chen Change Loy
Abstract: We introduce a new task -- language-driven video inpainting, which uses natural language instructions to guide the inpainting process. This approach overcomes the limitations of traditional video inpainting methods that depend on manually labeled binary masks, a process often tedious and labor-intensive. We present the Remove Objects from Videos by Instructions (ROVI) dataset, containing 5,650 videos and 9,091 inpainting results, to support training and evaluation for this task. We also propose a novel diffusion-based language-driven video inpainting framework, the first end-to-end baseline for this task, integrating Multimodal Large Language Models to understand and execute complex language-based inpainting requests effectively. Our comprehensive results showcase the dataset's versatility and the model's effectiveness in various language-instructed inpainting scenarios. We will make datasets, code, and models publicly available.
Authors: Xiangtai Li, Haobo Yuan, Wei Li, Henghui Ding, Size Wu, Wenwei Zhang, Yining Li, Kai Chen, Chen Change Loy
Abstract: In this work, we address various segmentation tasks, each traditionally tackled by distinct or partially unified models. We propose OMG-Seg, One Model that is Good enough to efficiently and effectively handle all the segmentation tasks, including image semantic, instance, and panoptic segmentation, as well as their video counterparts, open vocabulary settings, prompt-driven, interactive segmentation like SAM, and video object segmentation. To our knowledge, this is the first model to handle all these tasks in one model and achieve satisfactory performance. We show that OMG-Seg, a transformer-based encoder-decoder architecture with task-specific queries and outputs, can support over ten distinct segmentation tasks and yet significantly reduce computational and parameter overhead across various tasks and datasets. We rigorously evaluate the inter-task influences and correlations during co-training. Code and models are available at https://github.com/lxtGH/OMG-Seg.
Authors: Rohaifa Khaldi, Siham Tabik, Sergio Puertas-Ruiz, Julio Pe\~nas de Giles, Jos\'e Antonio H\'odar Correa, Regino Zamora, Domingo Alcaraz Segura
Abstract: Monitoring the distribution and size of long-living large shrubs, such as junipers, is crucial for assessing the long-term impacts of global change on high-mountain ecosystems. While deep learning models have shown remarkable success in object segmentation, adapting these models to detect shrub species with polymorphic nature remains challenging. In this research, we release a large dataset of individual shrub delineations on freely available satellite imagery and use an instance segmentation model to map all junipers over the treeline for an entire biosphere reserve (Sierra Nevada, Spain). To optimize performance, we introduced a novel dual data construction approach: using photo-interpreted (PI) data for model development and fieldwork (FW) data for validation. To account for the polymorphic nature of junipers during model evaluation, we developed a soft version of the Intersection over Union metric. Finally, we assessed the uncertainty of the resulting map in terms of canopy cover and density of shrubs per size class. Our model achieved an F1-score in shrub delineation of 87.87% on the PI data and 76.86% on the FW data. The R2 and RMSE of the observed versus predicted relationship were 0.63 and 6.67% for canopy cover, and 0.90 and 20.62 for shrub density. The greater density of larger shrubs in lower altitudes and smaller shrubs in higher altitudes observed in the model outputs was also present in the PI and FW data, suggesting an altitudinal uplift in the optimal performance of the species. This study demonstrates that deep learning applied on freely available high-resolution satellite imagery is useful to detect medium to large shrubs of high ecological value at the regional scale, which could be expanded to other high-mountains worldwide and to historical and forthcoming imagery.
Authors: Yongwei Nie, Changzhen Liu, Chengjiang Long, Qing Zhang, Guiqing Li, Hongmin Cai
Abstract: Besides a 3D mesh, Human Mesh Recovery (HMR) methods usually need to estimate a camera for computing 2D reprojection loss. Previous approaches may encounter the following problem: both the mesh and camera are not correct but the combination of them can yield a low reprojection loss. To alleviate this problem, we define multiple RoIs (region of interest) containing the same human and propose a multiple-RoI-based HMR method. Our key idea is that with multiple RoIs as input, we can estimate multiple local cameras and have the opportunity to design and apply additional constraints between cameras to improve the accuracy of the cameras and, in turn, the accuracy of the corresponding 3D mesh. To implement this idea, we propose a RoI-aware feature fusion network by which we estimate a 3D mesh shared by all RoIs as well as local cameras corresponding to the RoIs. We observe that local cameras can be converted to the camera of the full image through which we construct a local camera consistency loss as the additional constraint imposed on local cameras. Another benefit of introducing multiple RoIs is that we can encapsulate our network into a contrastive learning framework and apply a contrastive loss to regularize the training of our network. Experiments demonstrate the effectiveness of our multi-RoI HMR method and superiority to recent prior arts. Our code is available at https://github.com/CptDiaos/Multi-RoI.
Authors: Zidong Wang, Zeyu Lu, Di Huang, Cai Zhou, Wanli Ouyang, Lei Bai
Abstract: Nature is infinitely resolution-free. In the context of this reality, existing diffusion models, such as Diffusion Transformers, often face challenges when processing image resolutions outside of their trained domain. To address this limitation, we conceptualize images as sequences of tokens with dynamic sizes, rather than traditional methods that perceive images as fixed-resolution grids. This perspective enables a flexible training strategy that seamlessly accommodates various aspect ratios during both training and inference, thus promoting resolution generalization and eliminating biases introduced by image cropping. On this basis, we present the Flexible Vision Transformer (FiT), a transformer architecture specifically designed for generating images with unrestricted resolutions and aspect ratios. We further upgrade the FiT to FiTv2 with several innovative designs, includingthe Query-Key vector normalization, the AdaLN-LoRA module, a rectified flow scheduler, and a Logit-Normal sampler. Enhanced by a meticulously adjusted network structure, FiTv2 exhibits 2x convergence speed of FiT. When incorporating advanced training-free extrapolation techniques, FiTv2 demonstrates remarkable adaptability in both resolution extrapolation and diverse resolution generation. Additionally, our exploration of the scalability of the FiTv2 model reveals that larger models exhibit better computational efficiency. Furthermore, we introduce an efficient post-training strategy to adapt a pre-trained model for the high-resolution generation. Comprehensive experiments demonstrate the exceptional performance of FiTv2 across a broad range of resolutions. We have released all the codes and models at https://github.com/whlzy/FiT to promote the exploration of diffusion transformer models for arbitrary-resolution image generation.
Authors: Hao Wang, Sayed Pedram Haeri Boroujeni, Xiwen Chen, Ashish Bastola, Huayu Li, Wenhui Zhu, Abolfazl Razi
Abstract: Wildfires are a significant threat to ecosystems and human infrastructure, leading to widespread destruction and environmental degradation. Recent advancements in deep learning and generative models have enabled new methods for wildfire detection and monitoring. However, the scarcity of annotated wildfire images limits the development of robust models for these tasks. In this work, we present the FLAME Diffuser, a training-free, diffusion-based framework designed to generate realistic wildfire images with paired ground truth. Our framework uses augmented masks, sampled from real wildfire data, and applies Perlin noise to guide the generation of realistic flames. By controlling the placement of these elements within the image, we ensure precise integration while maintaining the original images style. We evaluate the generated images using normalized Frechet Inception Distance, CLIP Score, and a custom CLIP Confidence metric, demonstrating the high quality and realism of the synthesized wildfire images. Specifically, the fusion of Perlin noise in this work significantly improved the quality of synthesized images. The proposed method is particularly valuable for enhancing datasets used in downstream tasks such as wildfire detection and monitoring.
Authors: Kaiwen Cai, Zhekai Duan, Gaowen Liu, Charles Fleming, Chris Xiaoxuan Lu
Abstract: Recent advancements in Vision-Language (VL) models have sparked interest in their deployment on edge devices, yet challenges in handling diverse visual modalities, manual annotation, and computational constraints remain. We introduce EdgeVL, a novel framework that bridges this gap by seamlessly integrating dual-modality knowledge distillation and quantization-aware contrastive learning. This approach enables the adaptation of large VL models, like CLIP, for efficient use with both RGB and non-RGB images on resource-limited devices without the need for manual annotations. EdgeVL not only transfers visual language alignment capabilities to compact models but also maintains feature quality post-quantization, significantly enhancing open-vocabulary classification performance across various visual modalities. Our work represents the first systematic effort to adapt large VL models for edge deployment, showcasing up to 15.4% accuracy improvements on multiple datasets and up to 93-fold reduction in model size.
Authors: Novendra Setyawan, Ghufron Wahyu Kurniawan, Chi-Chia Sun, Jun-Wei Hsieh, Jing-Ming Guo, Wen-Kai Kuo
Abstract: Convolutional Neural Networks (CNNs) and Transformers have achieved remarkable success in computer vision tasks. However, their deep architectures often lead to high computational redundancy, making them less suitable for resource-constrained environments, such as edge devices. This paper introduces ParFormer, a novel vision transformer that addresses this challenge by incorporating a Parallel Mixer and a Sparse Channel Attention Patch Embedding (SCAPE). By combining convolutional and attention mechanisms, ParFormer improves feature extraction. This makes spatial feature extraction more efficient and cuts down on unnecessary computation. The SCAPE module further reduces computational redundancy while preserving essential feature information during down-sampling. Experimental results on the ImageNet-1K dataset show that ParFormer-T achieves 78.9\% Top-1 accuracy with a high throughput on a GPU that outperforms other small models with 2.56$\times$ higher throughput than MobileViT-S, 0.24\% faster than FasterNet-T2, and 1.79$\times$ higher than EdgeNeXt-S. For edge device deployment, ParFormer-T excels with a throughput of 278.1 images/sec, which is 1.38 $\times$ higher than EdgeNeXt-S and 2.36$\times$ higher than MobileViT-S, making it highly suitable for real-time applications in resource-constrained settings. The larger variant, ParFormer-L, reaches 83.5\% Top-1 accuracy, offering a balanced trade-off between accuracy and efficiency, surpassing many state-of-the-art models. In COCO object detection, ParFormer-M achieves 40.7 AP for object detection and 37.6 AP for instance segmentation, surpassing models like ResNet-50, PVT-S and PoolFormer-S24 with significantly higher efficiency. These results validate ParFormer as a highly efficient and scalable model for both high-performance and resource-constrained scenarios, making it an ideal solution for edge-based AI applications.
Authors: Tongkun Su, Jun Li, Xi Zhang, Haibo Jin, Hao Chen, Qiong Wang, Faqin Lv, Baoliang Zhao, Yin Hu
Abstract: Multimodal pre-training demonstrates its potential in the medical domain, which learns medical visual representations from paired medical reports. However, many pre-training tasks require extra annotations from clinicians, and most of them fail to explicitly guide the model to learn the desired features of different pathologies. In this paper, we utilize Visual Question Answering (VQA) for multimodal pre-training to guide the framework focusing on targeted pathological features. We leverage descriptions in medical reports to design multi-granular question-answer pairs associated with different diseases, which assist the framework in pre-training without requiring extra annotations from experts. We also propose a novel pre-training framework with a quasi-textual feature transformer, a module designed to transform visual features into a quasi-textual space closer to the textual domain via a contrastive learning strategy. This narrows the vision-language gap and facilitates modality alignment. Our framework is applied to four downstream tasks: report generation, classification, segmentation, and detection across five datasets. Extensive experiments demonstrate the superiority of our framework compared to other state-of-the-art methods. Our code is available at https://github.com/MoramiSu/QFT-MICCAI2024.
Authors: Xin Tie, Muheon Shin, Changhee Lee, Scott B. Perlman, Zachary Huemann, Amy J. Weisman, Sharon M. Castellino, Kara M. Kelly, Kathleen M. McCarten, Adina L. Alazraki, Junjie Hu, Steve Y. Cho, Tyler J. Bradshaw
Abstract: $\textbf{Purpose}$: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. $\textbf{Materials and Methods}$: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and $\Delta$SUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's $\rho$ correlations and employed bootstrap resampling for statistical analysis. $\textbf{Results}$: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, $\Delta$SUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's $\rho$ of 0.78, 0.80, 0.93 and 0.96, respectively. The performance remained high, with a slight decrease, in an external testing cohort. $\textbf{Conclusion}$: LAS-Net demonstrated significant improvements in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets.
Authors: Orcun Cetintas, Tim Meinhardt, Guillem Bras\'o, Laura Leal-Taix\'e
Abstract: Increasing the annotation efficiency of trajectory annotations from videos has the potential to enable the next generation of data-hungry tracking algorithms to thrive on large-scale datasets. Despite the importance of this task, there are currently very few works exploring how to efficiently label tracking datasets comprehensively. In this work, we introduce SPAM, a video label engine that provides high-quality labels with minimal human intervention. SPAM is built around two key insights: i) most tracking scenarios can be easily resolved. To take advantage of this, we utilize a pre-trained model to generate high-quality pseudo-labels, reserving human involvement for a smaller subset of more difficult instances; ii) handling the spatiotemporal dependencies of track annotations across time can be elegantly and efficiently formulated through graphs. Therefore, we use a unified graph formulation to address the annotation of both detections and identity association for tracks across time. Based on these insights, SPAM produces high-quality annotations with a fraction of ground truth labeling cost. We demonstrate that trackers trained on SPAM labels achieve comparable performance to those trained on human annotations while requiring only $3-20\%$ of the human labeling effort. Hence, SPAM paves the way towards highly efficient labeling of large-scale tracking datasets. We release all models and code.
Authors: Mohammad Areeb Qazi, Anees Ur Rehman Hashmi, Santosh Sanjeev, Ibrahim Almakky, Numan Saeed, Camila Gonzalez, Mohammad Yaqub
Abstract: Deep Learning has shown great success in reshaping medical imaging, yet it faces numerous challenges hindering widespread application. Issues like catastrophic forgetting and distribution shifts in the continuously evolving data stream increase the gap between research and applications. Continual Learning offers promise in addressing these hurdles by enabling the sequential acquisition of new knowledge without forgetting previous learnings in neural networks. In this survey, we comprehensively review the recent literature on continual learning in the medical domain, highlight recent trends, and point out the practical issues. Specifically, we survey the continual learning studies on classification, segmentation, detection, and other tasks in the medical domain. Furthermore, we develop a taxonomy for the reviewed studies, identify the challenges, and provide insights to overcome them. We also critically discuss the current state of continual learning in medical imaging, including identifying open problems and outlining promising future directions. We hope this survey will provide researchers with a useful overview of the developments in the field and will further increase interest in the community. To keep up with the fast-paced advancements in this field, we plan to routinely update the repository with the latest relevant papers at https://github.com/BioMedIA-MBZUAI/awesome-cl-in-medical .
URLs: https://github.com/BioMedIA-MBZUAI/awesome-cl-in-medical
Authors: Yasi Zhang, Peiyu Yu, Yaxuan Zhu, Yingshan Chang, Feng Gao, Ying Nian Wu, Oscar Leong
Abstract: Generative models based on flow matching have attracted significant attention for their simplicity and superior performance in high-resolution image synthesis. By leveraging the instantaneous change-of-variables formula, one can directly compute image likelihoods from a learned flow, making them enticing candidates as priors for downstream tasks such as inverse problems. In particular, a natural approach would be to incorporate such image probabilities in a maximum-a-posteriori (MAP) estimation problem. A major obstacle, however, lies in the slow computation of the log-likelihood, as it requires backpropagating through an ODE solver, which can be prohibitively slow for high-dimensional problems. In this work, we propose an iterative algorithm to approximate the MAP estimator efficiently to solve a variety of linear inverse problems. Our algorithm is mathematically justified by the observation that the MAP objective can be approximated by a sum of $N$ ``local MAP'' objectives, where $N$ is the number of function evaluations. By leveraging Tweedie's formula, we show that we can perform gradient steps to sequentially optimize these objectives. We validate our approach for various linear inverse problems, such as super-resolution, deblurring, inpainting, and compressed sensing, and demonstrate that we can outperform other methods based on flow matching.
Authors: Dixuan Lin, Yuxiang Zhang, Mengcheng Li, Yebin Liu, Wei Jing, Qi Yan, Qianying Wang, Hongwen Zhang
Abstract: In this paper, we introduce OmniHands, a universal approach to recovering interactive hand meshes and their relative movement from monocular or multi-view inputs. Our approach addresses two major limitations of previous methods: lacking a unified solution for handling various hand image inputs and neglecting the positional relationship of two hands within images. To overcome these challenges, we develop a universal architecture with novel tokenization and contextual feature fusion strategies, capable of adapting to a variety of tasks. Specifically, we propose a Relation-aware Two-Hand Tokenization (RAT) method to embed positional relation information into the hand tokens. In this way, our network can handle both single-hand and two-hand inputs and explicitly leverage relative hand positions, facilitating the reconstruction of intricate hand interactions in real-world scenarios. As such tokenization indicates the relative relationship of two hands, it also supports more effective feature fusion. To this end, we further develop a 4D Interaction Reasoning (FIR) module to fuse hand tokens in 4D with attention and decode them into 3D hand meshes and relative temporal movements. The efficacy of our approach is validated on several benchmark datasets. The results on in-the-wild videos and real-world scenarios demonstrate the superior performances of our approach for interactive hand reconstruction. More video results can be found on the project page: https://OmniHand.github.io.
Authors: Daniel Bogdoll, No\"el Ollick, Tim Joseph, Svetlana Pavlitska, J. Marius Z\"ollner
Abstract: Dealing with atypical traffic scenarios remains a challenging task in autonomous driving. However, most anomaly detection approaches cannot be trained on raw sensor data but require exposure to outlier data and powerful semantic segmentation models trained in a supervised fashion. This limits the representation of normality to labeled data, which does not scale well. In this work, we revisit unsupervised anomaly detection and present UMAD, leveraging generative world models and unsupervised image segmentation. Our method outperforms state-of-the-art unsupervised anomaly detection.
Authors: Daniel Bogdoll, Jan Imhof, Tim Joseph, Svetlana Pavlitska, J. Marius Z\"ollner
Abstract: In autonomous driving, the most challenging scenarios can only be detected within their temporal context. Most video anomaly detection approaches focus either on surveillance or traffic accidents, which are only a subfield of autonomous driving. We present HF$^2$-VAD$_{AD}$, a variation of the HF$^2$-VAD surveillance video anomaly detection method for autonomous driving. We learn a representation of normality from a vehicle's ego perspective and evaluate pixel-wise anomaly detections in rare and critical scenarios.
Authors: Tao Zhang, Xiangtai Li, Hao Fei, Haobo Yuan, Shengqiong Wu, Shunping Ji, Chen Change Loy, Shuicheng Yan
Abstract: Current universal segmentation methods demonstrate strong capabilities in pixel-level image and video understanding. However, they lack reasoning abilities and cannot be controlled via text instructions. In contrast, large vision-language multimodal models exhibit powerful vision-based conversation and reasoning capabilities but lack pixel-level understanding and have difficulty accepting visual prompts for flexible user interaction. This paper proposes OMG-LLaVA, a new and elegant framework combining powerful pixel-level vision understanding with reasoning abilities. It can accept various visual and text prompts for flexible user interaction. Specifically, we use a universal segmentation method as the visual encoder, integrating image information, perception priors, and visual prompts into visual tokens provided to the LLM. The LLM is responsible for understanding the user's text instructions and providing text responses and pixel-level segmentation results based on the visual information. We propose perception prior embedding to better integrate perception priors with image features. OMG-LLaVA achieves image-level, object-level, and pixel-level reasoning and understanding in a single model, matching or surpassing the performance of specialized methods on multiple benchmarks. Rather than using LLM to connect each specialist, our work aims at end-to-end training on one encoder, one decoder, and one LLM. The code and model have been released for further research.
Authors: Md Farhan Ishmam, Ishmam Tashdeed, Talukder Asir Saadat, Md Hamjajul Ashmafee, Abu Raihan Mostofa Kamal, Md. Azam Hossain
Abstract: Can Visual Question Answering (VQA) systems perform just as well when deployed in the real world? Or are they susceptible to realistic corruption effects e.g. image blur, which can be detrimental in sensitive applications, such as medical VQA? While linguistic or textual robustness has been thoroughly explored in the VQA literature, there has yet to be any significant work on the visual robustness of VQA models. We propose the first large-scale benchmark comprising 213,000 augmented images, challenging the visual robustness of multiple VQA models and assessing the strength of realistic visual corruptions. Additionally, we have designed several robustness evaluation metrics that can be aggregated into a unified metric and tailored to fit a variety of use cases. Our experiments reveal several insights into the relationships between model size, performance, and robustness with the visual corruptions. Our benchmark highlights the need for a balanced approach in model development that considers model performance without compromising the robustness.
Authors: Sushil Sharma, Arindam Das, Ganesh Sistu, Mark Halton, Ciar\'an Eising
Abstract: Predicting ego vehicle trajectories remains a critical challenge, especially in urban and dense areas due to the unpredictable behaviours of other vehicles and pedestrians. Multimodal trajectory prediction enhances decision-making by considering multiple possible future trajectories based on diverse sources of environmental data. In this approach, we leverage ResNet-50 to extract image features from high-definition map data and use IMU sensor data to calculate speed, acceleration, and yaw rate. A temporal probabilistic network is employed to compute potential trajectories, selecting the most accurate and highly probable trajectory paths. This method integrates HD map data to improve the robustness and reliability of trajectory predictions for autonomous vehicles.
Authors: Daniel Jakab, Alexander Braun, Cathaoir Agnew, Reenu Mohandas, Brian Michael Deegan, Dara Molloy, Enda Ward, Tony Scanlan, Ciar\'an Eising
Abstract: Automotive simulation can potentially compensate for a lack of training data in computer vision applications. However, there has been little to no image quality evaluation of automotive simulation and the impact of optical degradations on simulation is little explored. In this work, we investigate Virtual KITTI and the impact of applying variations of Gaussian blur on image sharpness. Furthermore, we consider object detection, a common computer vision application on three different state-of-the-art models, thus allowing us to characterize the relationship between object detection and sharpness. It was found that while image sharpness (MTF50) degrades from an average of 0.245cy/px to approximately 0.119cy/px; object detection performance stays largely robust within 0.58\%(Faster RCNN), 1.45\%(YOLOF) and 1.93\%(DETR) across all respective held-out test sets.
Authors: Anam Manzoor, Aryan Singh, Ganesh Sistu, Reenu Mohandas, Eoin Grua, Anthony Scanlan, Ciar\'an Eising
Abstract: This study investigates the effectiveness of modern Deformable Convolutional Neural Networks (DCNNs) for semantic segmentation tasks, particularly in autonomous driving scenarios with fisheye images. These images, providing a wide field of view, pose unique challenges for extracting spatial and geometric information due to dynamic changes in object attributes. Our experiments focus on segmenting the WoodScape fisheye image dataset into ten distinct classes, assessing the Deformable Networks' ability to capture intricate spatial relationships and improve segmentation accuracy. Additionally, we explore different loss functions to address class imbalance issues and compare the performance of conventional CNN architectures with Deformable Convolution-based CNNs, including Vanilla U-Net and Residual U-Net architectures. The significant improvement in mIoU score resulting from integrating Deformable CNNs demonstrates their effectiveness in handling the geometric distortions present in fisheye imagery, exceeding the performance of traditional CNN architectures. This underscores the significant role of Deformable convolution in enhancing semantic segmentation performance for fisheye imagery.
Authors: Hao Shu
Abstract: Image Edge detection (ED) is a base task in computer vision. While the performance of the ED algorithm has been improved greatly by introducing CNN-based models, current models still suffer from unsatisfactory precision rates especially when only a low error toleration distance is allowed. Therefore, model architecture for more precise predictions still needs an investigation. On the other hand, the unavoidable noise training data provided by humans would lead to unsatisfactory model predictions even when inputs are edge maps themselves, which also needs a solution. In this paper, more precise ED models are presented with cascaded skipping density blocks (CSDB). Our models obtain state-of-the-art(SOTA) predictions in several datasets, especially in average precision rate (AP), over a high-standard benchmark, which is confirmed by extensive experiments. Also, a novel modification on data augmentation for training is employed, which allows noiseless data to be employed in model training for the first time, and thus further improves the model performance. The relative Python codes can be found on https://github.com/Hao-B-Shu/SDPED.
Authors: Zhicheng Zou, Nantheera Anantrasirichai
Abstract: Atmospheric turbulence in long-range imaging significantly degrades the quality and fidelity of captured scenes due to random variations in both spatial and temporal dimensions. These distortions present a formidable challenge across various applications, from surveillance to astronomy, necessitating robust mitigation strategies. While model-based approaches achieve good results, they are very slow. Deep learning approaches show promise in image and video restoration but have struggled to address these spatiotemporal variant distortions effectively. This paper proposes a new framework that combines geometric restoration with an enhancement module. Random perturbations and geometric distortion are removed using a pyramid architecture with deformable 3D convolutions, resulting in aligned frames. These frames are then used to reconstruct a sharp, clear image via a multi-scale architecture of 3D Swin Transformers. The proposed framework demonstrates superior performance over the state of the art for both synthetic and real atmospheric turbulence effects, with reasonable speed and model size.
Authors: Susung Hong
Abstract: Conditional diffusion models have shown remarkable success in visual content generation, producing high-quality samples across various domains, largely due to classifier-free guidance (CFG). Recent attempts to extend guidance to unconditional models have relied on heuristic techniques, resulting in suboptimal generation quality and unintended effects. In this work, we propose Smoothed Energy Guidance (SEG), a novel training- and condition-free approach that leverages the energy-based perspective of the self-attention mechanism to enhance image generation. By defining the energy of self-attention, we introduce a method to reduce the curvature of the energy landscape of attention and use the output as the unconditional prediction. Practically, we control the curvature of the energy landscape by adjusting the Gaussian kernel parameter while keeping the guidance scale parameter fixed. Additionally, we present a query blurring method that is equivalent to blurring the entire attention weights without incurring quadratic complexity in the number of tokens. In our experiments, SEG achieves a Pareto improvement in both quality and the reduction of side effects. The code is available at https://github.com/SusungHong/SEG-SDXL.
Authors: Matan Atad, David Schinz, Hendrik Moeller, Robert Graf, Benedikt Wiestler, Daniel Rueckert, Nassir Navab, Jan S. Kirschke, Matthias Keicher
Abstract: Counterfactual explanations (CEs) aim to enhance the interpretability of machine learning models by illustrating how alterations in input features would affect the resulting predictions. Common CE approaches require an additional model and are typically constrained to binary counterfactuals. In contrast, we propose a novel method that operates directly on the latent space of a generative model, specifically a Diffusion Autoencoder (DAE). This approach offers inherent interpretability by enabling the generation of CEs and the continuous visualization of the model's internal representation across decision boundaries. Our method leverages the DAE's ability to encode images into a semantically rich latent space in an unsupervised manner, eliminating the need for labeled data or separate feature extraction models. We show that these latent representations are helpful for medical condition classification and the ordinal regression of severity pathologies, such as vertebral compression fractures (VCF) and diabetic retinopathy (DR). Beyond binary CEs, our method supports the visualization of ordinal CEs using a linear model, providing deeper insights into the model's decision-making process and enhancing interpretability. Experiments across various medical imaging datasets demonstrate the method's advantages in interpretability and versatility. The linear manifold of the DAE's latent space allows for meaningful interpolation and manipulation, making it a powerful tool for exploring medical image properties. Our code is available at https://doi.org/10.5281/zenodo.13859266.
Authors: Xiao Han, Yiming Ren, Yichen Yao, Yujing Sun, Yuexin Ma
Abstract: Human motion prediction is crucial for human-centric multimedia understanding and interacting. Current methods typically rely on ground truth human poses as observed input, which is not practical for real-world scenarios where only raw visual sensor data is available. To implement these methods in practice, a pre-phrase of pose estimation is essential. However, such two-stage approaches often lead to performance degradation due to the accumulation of errors. Moreover, reducing raw visual data to sparse keypoint representations significantly diminishes the density of information, resulting in the loss of fine-grained features. In this paper, we propose \textit{LiDAR-HMP}, the first single-LiDAR-based 3D human motion prediction approach, which receives the raw LiDAR point cloud as input and forecasts future 3D human poses directly. Building upon our novel structure-aware body feature descriptor, LiDAR-HMP adaptively maps the observed motion manifold to future poses and effectively models the spatial-temporal correlations of human motions for further refinement of prediction results. Extensive experiments show that our method achieves state-of-the-art performance on two public benchmarks and demonstrates remarkable robustness and efficacy in real-world deployments.
Authors: Jun Yan, Pengyu Wang, Danni Wang, Weiquan Huang, Daniel Watzenig, Huilin Yin
Abstract: Semantic segmentation is a significant perception task in autonomous driving. It suffers from the risks of adversarial examples. In the past few years, deep learning has gradually transitioned from convolutional neural network (CNN) models with a relatively small number of parameters to foundation models with a huge number of parameters. The segment-anything model (SAM) is a generalized image segmentation framework that is capable of handling various types of images and is able to recognize and segment arbitrary objects in an image without the need to train on a specific object. It is a unified model that can handle diverse downstream tasks, including semantic segmentation, object detection, and tracking. In the task of semantic segmentation for autonomous driving, it is significant to study the zero-shot adversarial robustness of SAM. Therefore, we deliver a systematic empirical study on the robustness of SAM without additional training. Based on the experimental results, the zero-shot adversarial robustness of the SAM under the black-box corruptions and white-box adversarial attacks is acceptable, even without the need for additional training. The finding of this study is insightful in that the gigantic model parameters and huge amounts of training data lead to the phenomenon of emergence, which builds a guarantee of adversarial robustness. SAM is a vision foundation model that can be regarded as an early prototype of an artificial general intelligence (AGI) pipeline. In such a pipeline, a unified model can handle diverse tasks. Therefore, this research not only inspects the impact of vision foundation models on safe autonomous driving but also provides a perspective on developing trustworthy AGI. The code is available at: https://github.com/momo1986/robust_sam_iv.
Authors: Jumabek Alikhanov, Dilshod Obidov, Hakil Kim
Abstract: The Lightweight Integrated Tracking-Feature Extraction (LITE) paradigm is introduced as a novel multi-object tracking (MOT) approach. It enhances ReID-based trackers by eliminating inference, pre-processing, post-processing, and ReID model training costs. LITE uses real-time appearance features without compromising speed. By integrating appearance feature extraction directly into the tracking pipeline using standard CNN-based detectors such as YOLOv8m, LITE demonstrates significant performance improvements. The simplest implementation of LITE on top of classic DeepSORT achieves a HOTA score of 43.03% at 28.3 FPS on the MOT17 benchmark, making it twice as fast as DeepSORT on MOT17 and four times faster on the more crowded MOT20 dataset, while maintaining similar accuracy. Additionally, a new evaluation framework for tracking-by-detection approaches reveals that conventional trackers like DeepSORT remain competitive with modern state-of-the-art trackers when evaluated under fair conditions. The code will be available post-publication at https://github.com/Jumabek/LITE.
Authors: Zhenhuan Liu, Shuai Liu, Zhiwei Ning, Jie Yang, Wei Liu
Abstract: We present CD-NGP, which is a fast and scalable representation for 3D reconstruction and novel view synthesis in dynamic scenes. Inspired by continual learning, our method first segments input videos into multiple chunks, followed by training the model chunk by chunk, and finally, fuses features of the first branch and subsequent branches. Experiments on the prevailing DyNeRF dataset demonstrate that our proposed novel representation reaches a great balance between memory consumption, model size, training speed, and rendering quality. Specifically, our method consumes $85\%$ less training memory ($<14$GB) than offline methods and requires significantly lower streaming bandwidth ($<0.4$MB/frame) than other online alternatives.
Authors: Georgios Pantazopoulos, Malvina Nikandrou, Alessandro Suglia, Oliver Lemon, Arash Eshghi
Abstract: This study explores replacing Transformers in Visual Language Models (VLMs) with Mamba, a recent structured state space model (SSM) that demonstrates promising performance in sequence modeling. We test models up to 3B parameters under controlled conditions, showing that Mamba-based VLMs outperforms Transformers-based VLMs in captioning, question answering, and reading comprehension. However, we find that Transformers achieve greater performance in visual grounding and the performance gap widens with scale. We explore two hypotheses to explain this phenomenon: 1) the effect of task-agnostic visual encoding on the updates of the hidden states, and 2) the difficulty in performing visual grounding from the perspective of in-context multimodal retrieval. Our results indicate that a task-aware encoding yields minimal performance gains on grounding, however, Transformers significantly outperform Mamba at in-context multimodal retrieval. Overall, Mamba shows promising performance on tasks where the correct output relies on a summary of the image but struggles when retrieval of explicit information from the context is required.
Authors: Zicheng Duan, Yuxuan Ding, Chenhui Gou, Ziqin Zhou, Ethan Smith, Lingqiao Liu
Abstract: Zero-shot subject-driven image generation aims to produce images that incorporate a subject from a given example image. The challenge lies in preserving the subject's identity while aligning with the text prompt which often requires modifying certain aspects of the subject's appearance. Despite advancements in diffusion model based methods, existing approaches still struggle to balance identity preservation with text prompt alignment. In this study, we conducted an in-depth investigation into this issue and uncovered key insights for achieving effective identity preservation while maintaining a strong balance. Our key findings include: (1) the design of the subject image encoder significantly impacts identity preservation quality, and (2) separating text and subject guidance is crucial for both text alignment and identity preservation. Building on these insights, we introduce a new approach called EZIGen, which employs two main strategies: a carefully crafted subject image Encoder based on the pretrained UNet of the Stable Diffusion model to ensure high-quality identity transfer, following a process that decouples the guidance stages and iteratively refines the initial image layout. Through these strategies, EZIGen achieves state-of-the-art results on multiple subject-driven benchmarks with a unified model and 100 times less training data. The demo page is available at: https://zichengduan.github.io/pages/EZIGen/index.html.
URLs: https://zichengduan.github.io/pages/EZIGen/index.html.
Authors: Hui Shen, Zhongwei Wan, Xin Wang, Mi Zhang
Abstract: Mamba and Vision Mamba (Vim) models have shown their potential as an alternative to methods based on Transformer architecture. This work introduces Fast Mamba for Vision (Famba-V), a cross-layer token fusion technique to enhance the training efficiency of Vim models. The key idea of Famba-V is to identify and fuse similar tokens across different Vim layers based on a suit of cross-layer strategies instead of simply applying token fusion uniformly across all the layers that existing works propose. We evaluate the performance of Famba-V on CIFAR-100. Our results show that Famba-V is able to enhance the training efficiency of Vim models by reducing both training time and peak memory usage during training. Moreover, the proposed cross-layer strategies allow Famba-V to deliver superior accuracy-efficiency trade-offs. These results all together demonstrate Famba-V as a promising efficiency enhancement technique for Vim models.
Authors: Neil De La Fuente, Mireia Maj\'o, Irina Luzko, Henry C\'ordova, Gloria Fern\'andez-Esparrach, Jorge Bernal
Abstract: Accurate and robust medical image classification is a challenging task, especially in application domains where available annotated datasets are small and present high imbalance between target classes. Considering that data acquisition is not always feasible, especially for underrepresented classes, our approach introduces a novel synthetic augmentation strategy using class-specific Variational Autoencoders (VAEs) and latent space interpolation to improve discrimination capabilities. By generating realistic, varied synthetic data that fills feature space gaps, we address issues of data scarcity and class imbalance. The method presented in this paper relies on the interpolation of latent representations within each class, thus enriching the training set and improving the model's generalizability and diagnostic accuracy. The proposed strategy was tested in a small dataset of 321 images created to train and validate an automatic method for assessing the quality of cleanliness of esophagogastroduodenoscopy images. By combining real and synthetic data, an increase of over 18\% in the accuracy of the most challenging underrepresented class was observed. The proposed strategy not only benefited the underrepresented class but also led to a general improvement in other metrics, including a 6\% increase in global accuracy and precision.
Authors: Youngsun Lim, Hojun Choi, Pin-Yu Chen, Hyunjung Shim
Abstract: Despite the impressive success of text-to-image (TTI) generation models, existing studies overlook the issue of whether these models accurately convey factual information. In this paper, we focus on the problem of image hallucination, where images created by generation models fail to faithfully depict factual content. To address this, we introduce I-HallA (Image Hallucination evaluation with Question Answering), a novel automated evaluation metric that measures the factuality of generated images through visual question answering (VQA). We also introduce I-HallA v1.0, a curated benchmark dataset for this purpose. As part of this process, we develop a pipeline that generates high-quality question-answer pairs using multiple GPT-4 Omni-based agents, with human judgments to ensure accuracy. Our evaluation protocols measure image hallucination by testing if images from existing text-to-image models can correctly respond to these questions. The I-HallA v1.0 dataset comprises 1.2K diverse image-text pairs across nine categories with 1,000 rigorously curated questions covering various compositional challenges. We evaluate five text-to-image models using I-HallA and reveal that these state-of-the-art models often fail to accurately convey factual information. Moreover, we validate the reliability of our metric by demonstrating a strong Spearman correlation (rho=0.95) with human judgments. We believe our benchmark dataset and metric can serve as a foundation for developing factually accurate text-to-image generation models.
Authors: Kartik Jindgar, Grace W. Lindsay
Abstract: In recent years, analysis of remote sensing data has benefited immensely from borrowing techniques from the broader field of computer vision, such as the use of shared models pre-trained on large and diverse datasets. However, satellite imagery has unique features that are not accounted for in traditional computer vision, such as the existence of multiple revisits of the same location. Here, we explore the best way to use revisits in the framework of fine-tuning pre-trained remote sensing models. We focus on an applied research question of relevance to climate change mitigation -- power substation segmentation -- that is representative of applied uses of pre-trained models more generally. Through extensive tests of different multi-temporal input schemes across diverse model architectures, we find that fusing representations from multiple revisits in the model latent space is superior to other methods of using revisits, including as a form of data augmentation. We also find that a SWIN Transformer-based architecture performs better than U-nets and ViT-based models. We verify the generality of our results on a separate building density estimation task.
Authors: Helin Cao, Sven Behnke
Abstract: Perception systems play a crucial role in autonomous driving, incorporating multiple sensors and corresponding computer vision algorithms. 3D LiDAR sensors are widely used to capture sparse point clouds of the vehicle's surroundings. However, such systems struggle to perceive occluded areas and gaps in the scene due to the sparsity of these point clouds and their lack of semantics. To address these challenges, Semantic Scene Completion (SSC) jointly predicts unobserved geometry and semantics in the scene given raw LiDAR measurements, aiming for a more complete scene representation. Building on promising results of diffusion models in image generation and super-resolution tasks, we propose their extension to SSC by implementing the noising and denoising diffusion processes in the point and semantic spaces individually. To control the generation, we employ semantic LiDAR point clouds as conditional input and design local and global regularization losses to stabilize the denoising process. We evaluate our approach on autonomous driving datasets and our approach outperforms the state-of-the-art for SSC.
Authors: Aagam Shah, Reimar Weissbach, David A. Griggs, A. John Hart, Elif Ertekin, Sameh Tawfick
Abstract: With the increasing adoption of metal additive manufacturing (AM), researchers and practitioners are turning to data-driven approaches to optimise printing conditions. Cross-sectional images of melt tracks provide valuable information for tuning process parameters, developing parameter scaling data, and identifying defects. Here we present an image segmentation neural network that automatically identifies and measures melt track dimensions from a cross-section image. We use a U-Net architecture to train on a data set of 62 pre-labelled images obtained from different labs, machines, and materials coupled with image augmentation. When neural network hyperparameters such as batch size and learning rate are properly tuned, the learned model shows an accuracy for classification of over 99% and an F1 score over 90%. The neural network exhibits robustness when tested on images captured by various users, printed on different machines, and acquired using different microscopes. A post-processing module extracts the height and width of the melt pool, and the wetting angles. We discuss opportunities to improve model performance and avenues for transfer learning, such as extension to other AM processes such as directed energy deposition.
Authors: Chengxin Zheng, Junzhong Ji, Yanzhao Shi, Xiaodan Zhang, Liangqiong Qu
Abstract: Brain CT report generation is significant to aid physicians in diagnosing cranial diseases. Recent studies concentrate on handling the consistency between visual and textual pathological features to improve the coherence of report. However, there exist some challenges: 1) Redundant visual representing: Massive irrelevant areas in 3D scans distract models from representing salient visual contexts. 2) Shifted semantic representing: Limited medical corpus causes difficulties for models to transfer the learned textual representations to generative layers. This study introduces a Pathological Clue-driven Representation Learning (PCRL) model to build cross-modal representations based on pathological clues and naturally adapt them for accurate report generation. Specifically, we construct pathological clues from perspectives of segmented regions, pathological entities, and report themes, to fully grasp visual pathological patterns and learn cross-modal feature representations. To adapt the representations for the text generation task, we bridge the gap between representation learning and report generation by using a unified large language model (LLM) with task-tailored instructions. These crafted instructions enable the LLM to be flexibly fine-tuned across tasks and smoothly transfer the semantic representation for report generation. Experiments demonstrate that our method outperforms previous methods and achieves SoTA performance. Our code is available at "https://github.com/Chauncey-Jheng/PCRL-MRG".
Authors: Jiahui Fan, Fujun Luan, Jian Yang, Milo\v{s} Ha\v{s}an, Beibei Wang
Abstract: 3D Gaussian Splatting (3DGS) has shown its impressive power in novel view synthesis. However, creating relightable 3D assets, especially for objects with ill-defined shapes (e.g., fur), is still a challenging task. For these scenes, the decomposition between the light, geometry, and material is more ambiguous, as neither the surface constraints nor the analytical shading model hold. To address this issue, we propose RNG, a novel representation of relightable neural Gaussians, enabling the relighting of objects with both hard surfaces or fluffy boundaries. We avoid any assumptions in the shading model but maintain feature vectors, which can be further decoded by an MLP into colors, in each Gaussian point. Following prior work, we utilize a point light to reduce the ambiguity and introduce a shadow-aware condition to the network. We additionally propose a depth refinement network to help the shadow computation under the 3DGS framework, leading to better shadow effects under point lights. Furthermore, to avoid the blurriness brought by the alpha-blending in 3DGS, we design a hybrid forward-deferred optimization strategy. As a result, we achieve about $20\times$ faster in training and about $600\times$ faster in rendering than prior work based on neural radiance fields, with $60$ frames per second on an RTX4090.
Authors: Kaihang Pan, Zhaoyu Fan, Juncheng Li, Qifan Yu, Hao Fei, Siliang Tang, Richang Hong, Hanwang Zhang, Qianru Sun
Abstract: The swift advancement in Multimodal LLMs (MLLMs) also presents significant challenges for effective knowledge editing. Current methods, including intrinsic knowledge editing and external knowledge resorting, each possess strengths and weaknesses, struggling to balance the desired properties of reliability, generality, and locality when applied to MLLMs. In this paper, we propose UniKE, a novel multimodal editing method that establishes a unified perspective and paradigm for intrinsic knowledge editing and external knowledge resorting. Both types of knowledge are conceptualized as vectorized key-value memories, with the corresponding editing processes resembling the assimilation and accommodation phases of human cognition, conducted at the same semantic levels. Within such a unified framework, we further promote knowledge collaboration by disentangling the knowledge representations into the semantic and truthfulness spaces. Extensive experiments validate the effectiveness of our method, which ensures that the post-edit MLLM simultaneously maintains excellent reliability, generality, and locality. The code for UniKE will be available at \url{https://github.com/beepkh/UniKE}.
Authors: Alessandro Fontanella, Grant Mair, Joanna Wardlaw, Emanuele Trucco, Amos Storkey
Abstract: Segmentation masks of pathological areas are useful in many medical applications, such as brain tumour and stroke management. Moreover, healthy counterfactuals of diseased images can be used to enhance radiologists' training files and to improve the interpretability of segmentation models. In this work, we present a weakly supervised method to generate a healthy version of a diseased image and then use it to obtain a pixel-wise anomaly map. To do so, we start by considering a saliency map that approximately covers the pathological areas, obtained with ACAT. Then, we propose a technique that allows to perform targeted modifications to these regions, while preserving the rest of the image. In particular, we employ a diffusion model trained on healthy samples and combine Denoising Diffusion Probabilistic Model (DDPM) and Denoising Diffusion Implicit Model (DDIM) at each step of the sampling process. DDPM is used to modify the areas affected by a lesion within the saliency map, while DDIM guarantees reconstruction of the normal anatomy outside of it. The two parts are also fused at each timestep, to guarantee the generation of a sample with a coherent appearance and a seamless transition between edited and unedited parts. We verify that when our method is applied to healthy samples, the input images are reconstructed without significant modifications. We compare our approach with alternative weakly supervised methods on the task of brain lesion segmentation, achieving the highest mean Dice and IoU scores among the models considered.
Authors: Luca Scimeca, Alexander Rubinstein, Damien Teney, Seong Joon Oh, Armand Mihai Nicolicioiu, Yoshua Bengio
Abstract: Spurious correlations in the data, where multiple cues are predictive of the target labels, often lead to a phenomenon known as shortcut learning, where a model relies on erroneous, easy-to-learn cues while ignoring reliable ones. In this work, we propose DiffDiv an ensemble diversification framework exploiting Diffusion Probabilistic Models (DPMs) to mitigate this form of bias. We show that at particular training intervals, DPMs can generate images with novel feature combinations, even when trained on samples displaying correlated input features. We leverage this crucial property to generate synthetic counterfactuals to increase model diversity via ensemble disagreement. We show that DPM-guided diversification is sufficient to remove dependence on shortcut cues, without a need for additional supervised signals. We further empirically quantify its efficacy on several diversification objectives, and finally show improved generalization and diversification on par with prior work that relies on auxiliary data collection.
Authors: Yafei Hu, Quanting Xie, Vidhi Jain, Jonathan Francis, Jay Patrikar, Nikhil Keetha, Seungchan Kim, Yaqi Xie, Tianyi Zhang, Hao-Shu Fang, Shibo Zhao, Shayegan Omidshafiei, Dong-Ki Kim, Ali-akbar Agha-mohammadi, Katia Sycara, Matthew Johnson-Roberson, Dhruv Batra, Xiaolong Wang, Sebastian Scherer, Chen Wang, Zsolt Kira, Fei Xia, Yonatan Bisk
Abstract: Building general-purpose robots that operate seamlessly in any environment, with any object, and utilizing various skills to complete diverse tasks has been a long-standing goal in Artificial Intelligence. However, as a community, we have been constraining most robotic systems by designing them for specific tasks, training them on specific datasets, and deploying them within specific environments. These systems require extensively-labeled data and task-specific models. When deployed in real-world scenarios, such systems face several generalization issues and struggle to remain robust to distribution shifts. Motivated by the impressive open-set performance and content generation capabilities of web-scale, large-capacity pre-trained models (i.e., foundation models) in research fields such as Natural Language Processing (NLP) and Computer Vision (CV), we devote this survey to exploring (i) how these existing foundation models from NLP and CV can be applied to the field of general-purpose robotics, and also exploring (ii) what a robotics-specific foundation model would look like. We begin by providing a generalized formulation of how foundation models are used in robotics, and the fundamental barriers to making generalist robots universally applicable. Next, we establish a taxonomy to discuss current work exploring ways to leverage existing foundation models for robotics and develop ones catered to robotics. Finally, we discuss key challenges and promising future directions in using foundation models for enabling general-purpose robotic systems. We encourage readers to view our living GitHub repository 2 of resources, including papers reviewed in this survey, as well as related projects and repositories for developing foundation models for robotics.
Authors: Yizhe Zhang, He Bai, Ruixiang Zhang, Jiatao Gu, Shuangfei Zhai, Josh Susskind, Navdeep Jaitly
Abstract: Vision-Language Models (VLMs) have recently demonstrated incredible strides on diverse vision language tasks. We dig into vision-based deductive reasoning, a more sophisticated but less explored realm, and find previously unexposed blindspots in the current SOTA VLMs. Specifically, we leverage Raven's Progressive Matrices (RPMs), to assess VLMs' abilities to perform multi-hop relational and deductive reasoning relying solely on visual clues. We perform comprehensive evaluations of several popular VLMs employing standard strategies such as in-context learning, self-consistency, and Chain-of-thoughts (CoT) on three diverse datasets, including the Mensa IQ test, IntelligenceTest, and RAVEN. The results reveal that despite the impressive capabilities of LLMs in text-based reasoning, we are still far from achieving comparable proficiency in visual deductive reasoning. We found that certain standard strategies that are effective when applied to LLMs do not seamlessly translate to the challenges presented by visual reasoning tasks. A detailed analysis reveals that VLMs struggle to solve these tasks mainly because they are unable to perceive and comprehend multiple, confounding abstract patterns in RPM examples.
Authors: Cory Efird, Alex Murphy, Joel Zylberberg, Alona Fyshe
Abstract: Prior work has offered evidence for functional localization in the brain; different anatomical regions preferentially activate for certain types of visual input. For example, the fusiform face area preferentially activates for visual stimuli that include a face. However, the spectrum of visual semantics is extensive, and only a few semantically-tuned patches of cortex have so far been identified in the human brain. Using a multimodal (natural language and image) neural network architecture (CLIP) we train a highly accurate contrastive model that maps brain responses during naturalistic image viewing to CLIP embeddings. We then use a novel adaptation of the DBSCAN clustering algorithm to cluster the parameters of these participant-specific contrastive models. This reveals what we call Shared Decodable Concepts (SDCs): clusters in CLIP space that are decodable from common sets of voxels across multiple participants. Examining the images most and least associated with each SDC cluster gives us additional insight into the semantic properties of each SDC. We note SDCs for previously reported visual features (e.g. orientation tuning in early visual cortex) as well as visual semantic concepts such as faces, places and bodies. In cases where our method finds multiple clusters for a visuo-semantic concept, the least associated images allow us to dissociate between confounding factors. For example, we discovered two clusters of food images, one driven by color, the other by shape. We also uncover previously unreported areas such as regions of extrastriate body area (EBA) tuned for legs/hands and sensitivity to numerosity in right intraparietal sulcus, and more. Thus, our contrastive-learning methodology better characterizes new and existing visuo-semantic representations in the brain by leveraging multimodal neural network representations and a novel adaptation of clustering algorithms.
Authors: Grigor Bezirganyan, Sana Sellami, Laure Berti-\'Equille, S\'ebastien Fournier
Abstract: Multimodal Deep Learning enhances decision-making by integrating diverse information sources, such as texts, images, audio, and videos. To develop trustworthy multimodal approaches, it is essential to understand how uncertainty impacts these models. We propose LUMA, a unique benchmark dataset, featuring audio, image, and textual data from 50 classes, for learning from uncertain and multimodal data. It extends the well-known CIFAR 10/100 dataset with audio samples extracted from three audio corpora, and text data generated using the Gemma-7B Large Language Model (LLM). The LUMA dataset enables the controlled injection of varying types and degrees of uncertainty to achieve and tailor specific experiments and benchmarking initiatives. LUMA is also available as a Python package including the functions for generating multiple variants of the dataset with controlling the diversity of the data, the amount of noise for each modality, and adding out-of-distribution samples. A baseline pre-trained model is also provided alongside three uncertainty quantification methods: Monte-Carlo Dropout, Deep Ensemble, and Reliable Conflictive Multi-View Learning. This comprehensive dataset and its benchmarking tools are intended to promote and support the development, evaluation, and benchmarking of trustworthy and robust multimodal deep learning approaches. We anticipate that the LUMA dataset will help the ICLR community to design more trustworthy and robust machine learning approaches for safety critical applications.
Authors: Lin Fan, Yafei Ou, Cenyang Zheng, Pengyu Dai, Tamotsu Kamishima, Masayuki Ikebe, Kenji Suzuki, Xun Gong
Abstract: Multi-modal fusion is crucial in medical data research, enabling a comprehensive understanding of diseases and improving diagnostic performance by combining diverse modalities. However, multi-modal fusion faces challenges, including capturing interactions between modalities, addressing missing modalities, handling erroneous modal information, and ensuring interpretability. Many existing researchers tend to design different solutions for these problems, often overlooking the commonalities among them. This paper proposes a novel multi-modal fusion framework that achieves adaptive adjustment over the weights of each modality by introducing the Modal-Domain Attention (MDA). It aims to facilitate the fusion of multi-modal information while allowing for the inclusion of missing modalities or intrinsic noise, thereby enhancing the representation of multi-modal data. We provide visualizations of accuracy changes and MDA weights by observing the process of modal fusion, offering a comprehensive analysis of its interpretability. Extensive experiments on various gastrointestinal disease benchmarks, the proposed MDA maintains high accuracy even in the presence of missing modalities and intrinsic noise. One thing worth mentioning is that the visualization of MDA is highly consistent with the conclusions of existing clinical studies on the dependence of different diseases on various modalities. Code and dataset will be made available.
Authors: Vahid Jebraeeli, Bo Jiang, Hamid Krim, Derya Cansever
Abstract: Limited data availability in machine learning significantly impacts performance and generalization. Traditional augmentation methods enhance moderately sufficient datasets. GANs struggle with convergence when generating diverse samples. Diffusion models, while effective, have high computational costs. We introduce an Expansive Synthesis model generating large-scale, information-rich datasets from minimal samples. It uses expander graph mappings and feature interpolation to preserve data distribution and feature relationships. The model leverages neural networks' non-linear latent space, captured by a Koopman operator, to create a linear feature space for dataset expansion. An autoencoder with self-attention layers and optimal transport refines distributional consistency. We validate by comparing classifiers trained on generated data to those trained on original datasets. Results show comparable performance, demonstrating the model's potential to augment training data effectively. This work advances data generation, addressing scarcity in machine learning applications.
Authors: Gabriele Lozupone, Alessandro Bria, Francesco Fontanella, Frederick J. A. Meijer, Claudio De Stefano
Abstract: This study presents an innovative method for Alzheimer's disease diagnosis using 3D MRI designed to enhance the explainability of model decisions. Our approach adopts a soft attention mechanism, enabling 2D CNNs to extract volumetric representations. At the same time, the importance of each slice in decision-making is learned, allowing the generation of a voxel-level attention map to produce an explainable MRI. To test our method and ensure the reproducibility of our results, we chose a standardized collection of MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). On this dataset, our method significantly outperforms state-of-the-art methods in (i) distinguishing AD from cognitive normal (CN) with an accuracy of 0.856 and Matthew's correlation coefficient (MCC) of 0.712, representing improvements of 2.4% and 5.3% respectively over the second-best, and (ii) in the prognostic task of discerning stable from progressive mild cognitive impairment (MCI) with an accuracy of 0.725 and MCC of 0.443, showing improvements of 10.2% and 20.5% respectively over the second-best. We achieved this prognostic result by adopting a double transfer learning strategy, which enhanced sensitivity to morphological changes and facilitated early-stage AD detection. With voxel-level precision, our method identified which specific areas are being paid attention to, identifying these predominant brain regions: the hippocampus, the amygdala, the parahippocampal, and the inferior lateral ventricles. All these areas are clinically associated with AD development. Furthermore, our approach consistently found the same AD-related areas across different cross-validation folds, proving its robustness and precision in highlighting areas that align closely with known pathological markers of the disease.
Authors: Seamie Hayes, Sushil Sharma, Ciar\'an Eising
Abstract: Fusing different sensor modalities can be a difficult task, particularly if they are asynchronous. Asynchronisation may arise due to long processing times or improper synchronisation during calibration, and there must exist a way to still utilise this previous information for the purpose of safe driving, and object detection in ego vehicle/ multi-agent trajectory prediction. Difficulties arise in the fact that the sensor modalities have captured information at different times and also at different positions in space. Therefore, they are not spatially nor temporally aligned. This paper will investigate the challenge of radar and LiDAR sensors being asynchronous relative to the camera sensors, for various time latencies. The spatial alignment will be resolved before lifting into BEV space via the transformation of the radar/LiDAR point clouds into the new ego frame coordinate system. Only after this can we concatenate the radar/LiDAR point cloud and lifted camera features. Temporal alignment will be remedied for radar data only, we will implement a novel method of inferring the future radar point positions using the velocity information. Our approach to resolving the issue of sensor asynchrony yields promising results. We demonstrate velocity information can drastically improve IoU for asynchronous datasets, as for a time latency of 360 milliseconds (ms), IoU improves from 49.54 to 53.63. Additionally, for a time latency of 550ms, the camera+radar (C+R) model outperforms the camera+LiDAR (C+L) model by 0.18 IoU. This is an advancement in utilising the often-neglected radar sensor modality, which is less favoured than LiDAR for autonomous driving purposes.
Authors: Linkai Peng, Zheyuan Zhang, Gorkem Durak, Frank H. Miller, Alpay Medetalibeyoglu, Michael B. Wallace, Ulas Bagci
Abstract: Pancreatic cancer remains one of the leading causes of cancer-related mortality worldwide. Precise segmentation of pancreatic tumors from medical images is a bottleneck for effective clinical decision-making. However, achieving a high accuracy is often limited by the small size and availability of real patient data for training deep learning models. Recent approaches have employed synthetic data generation to augment training datasets. While promising, these methods may not yet meet the performance benchmarks required for real-world clinical use. This study critically evaluates the limitations of existing generative-AI based frameworks for pancreatic tumor segmentation. We conduct a series of experiments to investigate the impact of synthetic \textit{tumor size} and \textit{boundary definition} precision on model performance. Our findings demonstrate that: (1) strategically selecting a combination of synthetic tumor sizes is crucial for optimal segmentation outcomes, and (2) generating synthetic tumors with precise boundaries significantly improves model accuracy. These insights highlight the importance of utilizing refined synthetic data augmentation for enhancing the clinical utility of segmentation models in pancreatic cancer decision making including diagnosis, prognosis, and treatment plans. Our code will be available at https://github.com/lkpengcs/SynTumorAnalyzer.
Authors: Pengcheng Chen, Jin Ye, Guoan Wang, Yanjun Li, Zhongying Deng, Wei Li, Tianbin Li, Haodong Duan, Ziyan Huang, Yanzhou Su, Benyou Wang, Shaoting Zhang, Bin Fu, Jianfei Cai, Bohan Zhuang, Eric J Seibel, Junjun He, Yu Qiao
Abstract: Large Vision-Language Models (LVLMs) are capable of handling diverse data types such as imaging, text, and physiological signals, and can be applied in various fields. In the medical field, LVLMs have a high potential to offer substantial assistance for diagnosis and treatment. Before that, it is crucial to develop benchmarks to evaluate LVLMs' effectiveness in various medical applications. Current benchmarks are often built upon specific academic literature, mainly focusing on a single domain, and lacking varying perceptual granularities. Thus, they face specific challenges, including limited clinical relevance, incomplete evaluations, and insufficient guidance for interactive LVLMs. To address these limitations, we developed the GMAI-MMBench, the most comprehensive general medical AI benchmark with well-categorized data structure and multi-perceptual granularity to date. It is constructed from 284 datasets across 38 medical image modalities, 18 clinical-related tasks, 18 departments, and 4 perceptual granularities in a Visual Question Answering (VQA) format. Additionally, we implemented a lexical tree structure that allows users to customize evaluation tasks, accommodating various assessment needs and substantially supporting medical AI research and applications. We evaluated 50 LVLMs, and the results show that even the advanced GPT-4o only achieves an accuracy of 53.96%, indicating significant room for improvement. Moreover, we identified five key insufficiencies in current cutting-edge LVLMs that need to be addressed to advance the development of better medical applications. We believe that GMAI-MMBench will stimulate the community to build the next generation of LVLMs toward GMAI.
Authors: Lucas Nedel Kirsten, Zhicheng Fu, Nikhil Ambha Madhusudhana
Abstract: Recent advances in camera design and imaging technology have enabled the capture of high-quality images using smartphones. However, due to the limited dynamic range of digital cameras, the quality of photographs captured in environments with highly imbalanced lighting often results in poor-quality images. To address this issue, most devices capture multi-exposure frames and then use some multi-exposure fusion method to merge those frames into a final fused image. Nevertheless, most traditional and current deep learning approaches are unsuitable for real-time applications on mobile devices due to their heavy computational and memory requirements. We propose a new method for multi-exposure fusion based on an encoder-decoder deep learning architecture with efficient building blocks tailored for mobile devices. This efficient design makes our model capable of processing 4K resolution images in less than 2 seconds on mid-range smartphones. Our method outperforms state-of-the-art techniques regarding full-reference quality measures and computational efficiency (runtime and memory usage), making it ideal for real-time applications on hardware-constrained devices. Our code is available at: https://github.com/LucasKirsten/MobileMEF.
Authors: Doyoung Park, Jinsoo Kim, Qi Chang, Shuang Leng, Liang Zhong, Lohendran Baskaran
Abstract: The Agatston score, which is the sum of the calcification in the four main coronary arteries, has been widely used in the diagnosis of coronary artery disease (CAD). However, many studies have emphasized the importance of the vessel-specific Agatston score, as calcification in a specific vessel is significantly correlated with the occurrence of coronary heart disease (CHD). In this paper, we propose the Residual-block Inspired Coordinate Attention U-Net (RICAU-Net), which incorporates coordinate attention in two distinct manners and a customized combo loss function for lesion-specific coronary artery calcium (CAC) segmentation. This approach aims to tackle the high class-imbalance issue associated with small and sparse CAC lesions. Experimental results and the ablation study demonstrate that the proposed method outperforms the four other U-Net based methods used in medical applications, by achieving the highest per-lesion Dice scores across all four lesions.
Authors: Sven L\"upke, Yousef Yeganeh, Ehsan Adeli, Nassir Navab, Azade Farshad
Abstract: Recent advances in generative models for medical imaging have shown promise in representing multiple modalities. However, the variability in modality availability across datasets limits the general applicability of the synthetic data they produce. To address this, we present a novel physics-informed generative model capable of synthesizing a variable number of brain MRI modalities, including those not present in the original dataset. Our approach utilizes latent diffusion models and a two-step generative process: first, unobserved physical tissue property maps are synthesized using a latent diffusion model, and then these maps are combined with a physical signal model to generate the final MRI scan. Our experiments demonstrate the efficacy of this approach in generating unseen MR contrasts and preserving physical plausibility. Furthermore, we validate the distributions of generated tissue properties by comparing them to those measured in real brain tissue.
Authors: Guokang Wang, Hang Li, Shuyuan Zhang, Yanhong Liu, Huaping Liu
Abstract: In real-world scenarios, many robotic manipulation tasks are hindered by occlusions and limited fields of view, posing significant challenges for passive observation-based models that rely on fixed or wrist-mounted cameras. In this paper, we investigate the problem of robotic manipulation under limited visual observation and propose a task-driven asynchronous active vision-action model.Our model serially connects a camera Next-Best-View (NBV) policy with a gripper Next-Best Pose (NBP) policy, and trains them in a sensor-motor coordination framework using few-shot reinforcement learning. This approach allows the agent to adjust a third-person camera to actively observe the environment based on the task goal, and subsequently infer the appropriate manipulation actions.We trained and evaluated our model on 8 viewpoint-constrained tasks in RLBench. The results demonstrate that our model consistently outperforms baseline algorithms, showcasing its effectiveness in handling visual constraints in manipulation tasks.
Authors: Sunny Gupta, Mohit Jindal, Pankhi Kashyap, Pranav Jeevan, Amit Sethi
Abstract: Federated learning faces a critical challenge in balancing communication efficiency with rapid convergence, especially for second-order methods. While Newton-type algorithms achieve linear convergence in communication rounds, transmitting full Hessian matrices is often impractical due to quadratic complexity. We introduce Federated Learning with Enhanced Nesterov-Newton Sketch (FLeNS), a novel method that harnesses both the acceleration capabilities of Nesterov's method and the dimensionality reduction benefits of Hessian sketching. FLeNS approximates the centralized Newton's method without relying on the exact Hessian, significantly reducing communication overhead. By combining Nesterov's acceleration with adaptive Hessian sketching, FLeNS preserves crucial second-order information while preserving the rapid convergence characteristics. Our theoretical analysis, grounded in statistical learning, demonstrates that FLeNS achieves super-linear convergence rates in communication rounds - a notable advancement in federated optimization. We provide rigorous convergence guarantees and characterize tradeoffs between acceleration, sketch size, and convergence speed. Extensive empirical evaluation validates our theoretical findings, showcasing FLeNS's state-of-the-art performance with reduced communication requirements, particularly in privacy-sensitive and edge-computing scenarios. The code is available at https://github.com/sunnyinAI/FLeNS
Authors: Jiaheng Hu, Rose Hendrix, Ali Farhadi, Aniruddha Kembhavi, Roberto Martin-Martin, Peter Stone, Kuo-Hao Zeng, Kiana Ehsani
Abstract: In recent years, the Robotics field has initiated several efforts toward building generalist robot policies through large-scale multi-task Behavior Cloning. However, direct deployments of these policies have led to unsatisfactory performance, where the policy struggles with unseen states and tasks. How can we break through the performance plateau of these models and elevate their capabilities to new heights? In this paper, we propose FLaRe, a large-scale Reinforcement Learning fine-tuning framework that integrates robust pre-trained representations, large-scale training, and gradient stabilization techniques. Our method aligns pre-trained policies towards task completion, achieving state-of-the-art (SoTA) performance both on previously demonstrated and on entirely novel tasks and embodiments. Specifically, on a set of long-horizon mobile manipulation tasks, FLaRe achieves an average success rate of 79.5% in unseen environments, with absolute improvements of +23.6% in simulation and +30.7% on real robots over prior SoTA methods. By utilizing only sparse rewards, our approach can enable generalizing to new capabilities beyond the pretraining data with minimal human effort. Moreover, we demonstrate rapid adaptation to new embodiments and behaviors with less than a day of fine-tuning. Videos can be found on the project website at https://robot-flare.github.io/
Authors: Apoorva Vashisth, Dipam Patel, Damon Conover, Aniket Bera
Abstract: Autonomous robots are being employed in several mapping and data collection tasks due to their efficiency and low labor costs. In these tasks, the robots are required to map targets-of-interest in an unknown environment while constrained to a given resource budget such as path length or mission time. This is a challenging problem as each robot has to not only detect and avoid collisions from static obstacles in the environment but also has to model other robots' trajectories to avoid inter-robot collisions. We propose a novel deep reinforcement learning approach for multi-robot informative path planning to map targets-of-interest in an unknown 3D environment. A key aspect of our approach is an augmented graph that models other robots' trajectories to enable planning for communication and inter-robot collision avoidance. We train our decentralized reinforcement learning policy via the centralized training and decentralized execution paradigm. Once trained, our policy is also scalable to varying number of robots and does not require re-training. Our approach outperforms other state-of-the-art multi-robot target mapping approaches by 33.75% in terms of the number of discovered targets-of-interest. We open-source our code and model at: https://github.com/AccGen99/marl_ipp
Authors: Ziquan Wei, Tingting Dan, Jiaqi Ding, Guorong Wu
Abstract: Although modern imaging technologies allow us to study connectivity between two distinct brain regions in-vivo, an in-depth understanding of how anatomical structure supports brain function and how spontaneous functional fluctuations emerge remarkable cognition is still elusive. Meanwhile, tremendous efforts have been made in the realm of machine learning to establish the nonlinear mapping between neuroimaging data and phenotypic traits. However, the absence of neuroscience insight in the current approaches poses significant challenges in understanding cognitive behavior from transient neural activities. To address this challenge, we put the spotlight on the coupling mechanism of structural connectivity (SC) and functional connectivity (FC) by formulating such network neuroscience question into an expressive graph representation learning problem for high-order topology. Specifically, we introduce the concept of topological detour to characterize how a ubiquitous instance of FC (direct link) is supported by neural pathways (detour) physically wired by SC, which forms a cyclic loop interacted by brain structure and function. In the clich\'e of machine learning, the multi-hop detour pathway underlying SC-FC coupling allows us to devise a novel multi-head self-attention mechanism within Transformer to capture multi-modal feature representation from paired graphs of SC and FC. Taken together, we propose a biological-inspired deep model, coined as NeuroPath, to find putative connectomic feature representations from the unprecedented amount of neuroimages, which can be plugged into various downstream applications such as task recognition and disease diagnosis. We have evaluated NeuroPath on large-scale public datasets including HCP and UK Biobank under supervised and zero-shot learning, where the state-of-the-art performance by our NeuroPath indicates great potential in network neuroscience.
Authors: Qing Lyu, Jin Young Kim, Jeongchul Kim, Christopher T Whitlow
Abstract: Beta-amyloid positron emission tomography (A$\beta$-PET) imaging has become a critical tool in Alzheimer's disease (AD) research and diagnosis, providing insights into the pathological accumulation of amyloid plaques, one of the hallmarks of AD. However, the high cost, limited availability, and exposure to radioactivity restrict the widespread use of A$\beta$-PET imaging, leading to a scarcity of comprehensive datasets. Previous studies have suggested that structural magnetic resonance imaging (MRI), which is more readily available, may serve as a viable alternative for synthesizing A$\beta$-PET images. In this study, we propose an approach to utilize 3D diffusion models to synthesize A$\beta$-PET images from T1-weighted MRI scans, aiming to overcome the limitations associated with direct PET imaging. Our method generates high-quality A$\beta$-PET images for cognitive normal cases, although it is less effective for mild cognitive impairment (MCI) patients due to the variability in A$\beta$ deposition patterns among subjects. Our preliminary results suggest that incorporating additional data, such as a larger sample of MCI cases and multi-modality information including clinical and demographic details, cognitive and functional assessments, and longitudinal data, may be necessary to improve A$\beta$-PET image synthesis for MCI patients.
Authors: Osama Mustafa
Abstract: The application of deep learning in cancer research, particularly in early diagnosis, case understanding, and treatment strategy design, emphasizes the need for high-quality data. Generative AI, especially Generative Adversarial Networks (GANs), has emerged as a leading solution to challenges like class imbalance, robust learning, and model training, while addressing issues stemming from patient privacy and the scarcity of real data. Despite their promise, GANs face several challenges, both inherent and specific to histopathology data. Inherent issues include training imbalance, mode collapse, linear learning from insufficient discriminator feedback, and hard boundary convergence due to stringent feedback. Histopathology data presents a unique challenge with its complex representation, high spatial resolution, and multiscale features. To address these challenges, we propose a framework consisting of two components. First, we introduce a contrastive learning-based Multistage Progressive Finetuning Siamese Neural Network (MFT-SNN) for assessing the similarity between histopathology patches. Second, we implement a Reinforcement Learning-based External Optimizer (RL-EO) within the GAN training loop, serving as a reward signal generator. The modified discriminator loss function incorporates a weighted reward, guiding the GAN to maximize this reward while minimizing loss. This approach offers an external optimization guide to the discriminator, preventing generator overfitting and ensuring smooth convergence. Our proposed solution has been benchmarked against state-of-the-art (SOTA) GANs and a Denoising Diffusion Probabilistic model, outperforming previous SOTA across various metrics, including FID score, KID score, Perceptual Path Length, and downstream classification tasks.