检索结果-内蒙古大学图书馆

Commonsense Prototype for Outdoor Unsupervised 3D Object Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Hai Zhao, Shijia Huang, Xun Wen, Chenglu Li, Xin Wang, Cheng Fujian Key Laboratory of Sensing and Computing for Smart Cities Xiamen University China Section of Visual Computing and Interactive Media Texas A&M University United States

The prevalent approaches of unsupervised 3D object detection follow cluster-based pseudo-label generation and iterative self-training processes. However, the challenge arises due to the sparsity of LiDAR scans, which leads to pseudo-labels with erroneous size and position, resulting in subpar detection performance. To tackle this problem, this paper introduces a Commonsense Prototype-based Detector, termed CPD, for unsupervised 3D object detection. CPD first constructs Commonsense Prototype (CProto) characterized by high-quality bounding box and dense points, based on commonsense intuition. Subsequently, CPD refines the low-quality pseudo-labels by leveraging the size prior from CProto. Furthermore, CPD enhances the detection accuracy of sparsely scanned objects by the geometric knowledge from CProto. CPD outperforms state-of-the-art unsupervised 3D detectors on Waymo Open Dataset (WOD), PandaSet, and KITTI datasets by a large margin. Besides, by training CPD on WOD and testing on KITTI, CPD attains 90.85% and 81.01% 3D Average Precision on easy and moderate car classes, respectively. These achievements position CPD in close proximity to fully supervised detectors, highlighting the significance of our method. The code will be available at https://***/hailanyi/CPD. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

HINTED: Hard Instance Enhanced Detector with Mixed-Density Feature Fusion for Sparsely-Supervised 3D Object Detection

学校读者我要写书评

暂无评论

HINTED: Hard Instance Enhanced Detector with Mixed-Density F...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Qiming Xia Wei Ye Hai Wu Shijia Zhao Leyuan Xing Xun Huang Jinhao Deng Xin Li Chenglu Wen Cheng Wang Fujian Key Laboratory of Sensing and Computing for Smart Cities Xiamen University Xiamen China Section of Visual Computing and Interactive Media Texas A&M University Texas USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Current sparsely-supervised object detection methods largely depend on high threshold settings to derive high-quality pseudo labels from detector predictions. However, hard instances within point clouds frequently display incomplete structures, causing decreased confidence scores in their assigned pseudo-labels. Previous methods inevitably result in inadequate positive supervision for these instances. To address this problem, we propose a novel Hard INsTance Enhanced Detector (HINTED), for sparsely-supervised 3D object detection. Firstly, we design a self-boosting teacher (SBT) model to generate more potential pseudo-labels, enhancing the effectiveness of information transfer. Then, we introduce a mixed-density student (MDS) model to concentrate on hard instances during the training phase, thereby improving detection accuracy. Our extensive experiments on the KITTI dataset validate our method's superior performance. Compared with leading sparsely-supervised methods, HINTED significantly improves the detection performance on hard instances, no-tably outperforming fully-supervised methods in detecting challenging categories like cyclists. HINTED also significantly outperforms the state-of-the-art semi-supervised method on challenging categories. The code is available at https://***/xmuqimingxia/HINTED.

关键词： Training Point cloud compression Computer vision Three-dimensional displays Codes Accuracy Object detection

Spectrum-guided Feature Enhancement Network for Event Person Re-Identification

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tan, Hongchen Zhang, Yi Liu, Xiuping Yin, Baocai Ma, Nan Li, Xin Lu, Huchuan Beijing Institute of Artificial Intelligence Beijing University of Technology China Dalian University of Technology China Section of Visual Computing and Creative Technology School of Performance Texas A & M University United States

As a cutting-edge biosensor, the event camera holds significant potential in the field of computer vision, particularly regarding privacy preservation. However, compared to traditional cameras, event streams often contain noise and possess extremely sparse semantics, posing a formidable challenge for event-based person re-identification (event ReID). To address this, we introduce a novel event person re-identification network: the Spectrum-guided Feature Enhancement Network (SFE-Net). This network consists of two innovative components: the Multi-grain Spectrum Attention Mechanism (MSAM) and the Consecutive Patch Dropout Module (CPDM). MSAM employs a fourier spectrum transform strategy to filter event noise, while also utilizing an event-guided multi-granularity attention strategy to enhance and capture discriminative person semantics. CPDM employs a consecutive patch dropout strategy to generate multiple incomplete feature maps, encouraging the deep ReID model to equally perceive each effective region of the person’s body and capture robust person descriptors. Extensive experiments on Event Re-ID datasets demonstrate that our SFE-Net achieves the best performance in this task. © 2024, CC BY.

关键词： Cameras

Realistic electric field characterization of clinically used deformable large TMS coils in a large cohort

学校读者我要写书评

暂无评论

Brain Stimulation 2025年第4期18卷 1174-1183页

作者： Worbs, Torge Rumi, Bianka Madsen, Kristoffer H. Thielscher, Axel Section for Magnetic Resonance DTU Health Tech Technical University of Denmark Kgs Lyngby Denmark Danish Research Centre for Magnetic Resonance Department of Radiology and Nuclear Medicine Copenhagen University Hospital Amager and Hvidovre Copenhagen Denmark Section for Visual Computing Department of Applied Mathematics and Computer Science Technical University of Denmark Kgs Lyngby Denmark Sino-Danish College University of Chinese Academy of Sciences Beijing 100190 China

Background: Transcranial Magnetic Stimulation (TMS) therapies use both focal and unfocal coil designs. Unfocal designs often employ bendable windings and moveable parts, making realistic simulations of their electric fields in inter-individually varying head sizes and shapes challenging. This hampers comparisons of the various coil designs and prevents systematic evaluations of their dose-response relationships. Objective: Introduce and validate a novel method for optimizing the position and shape of flexible coils taking individual head anatomies into account. Evaluate the impact of realistic modeling of flexible coils on the electric field simulated in the brain. Methods: Accurate models of four coils (Brainsway H1, H4, H7;MagVenture MST-Twin) were derived from computed tomography data and mechanical measurements. A generic representation of coil deformations by concatenated linear transformations was introduced and validated. This served as basis for a principled approach to optimize the coil positions and shapes, and to optionally maximize the electric field strength in a region of interest (ROI). Results: For all four coil models, the new method achieved configurations that followed the scalp anatomy while robustly preventing coil-scalp intersections on 1100 head models. In contrast, setting only the coil center positions without shape deformation regularly led to physically impossible configurations. This also affected the electric field calculated in the cortex, with a median peak difference of ∼16 %. In addition, the new method outperformed grid search-based optimization for maximizing the electric field of a standard Fig. 8 coil in a ROI with a comparable computational complexity. Conclusion: Our approach alleviates practical hurdles that so far hampered accurate simulations of bendable coils. This enables systematic comparison of dose-response relationships across the various coil designs employed in therapy. © 2025

关键词： Brainsway H-Coils Electric field calculations MagVenture MST-Twin coil Personalized dosing TMS

Next-Future: Sample-Efficient Policy Learning for Robotic-Arm Tasks

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Özgür, Fikrican Zurbrügg, René Kumar, Suryansh ETH Zürich Switzerland RSL Group ETH Zürich Switzerland Visual Computing and Computational Media Section College of PVFA Department of Electric and Computer Engineering Department of Computer Science and Engineering Texas A&M University College StationTX United States

Hindsight Experience Replay (HER) is widely regarded as the state-of-the-art algorithm for achieving sample-efficient multi-goal reinforcement learning (RL) in robotic manipulation tasks with binary rewards. HER facilitates learning from failed attempts by replaying trajectories with redefined goals. However, it relies on a heuristic-based replay method that lacks a principled framework. To address this limitation, we introduce a novel replay strategy, "Next-Future", which focuses on rewarding single-step transitions. This approach significantly enhances sample efficiency and accuracy in learning multi-goal Markov decision processes (MDPs), particularly under stringent accuracy requirements—a critical aspect for performing complex and precise robotic-arm tasks. We demonstrate the efficacy of our method by highlighting how single-step learning enables improved value approximation within the multi-goal RL framework. The performance of the proposed replay strategy is evaluated across eight challenging robotic manipulation tasks, using ten random seeds for training. Our results indicate substantial improvements in sample efficiency for seven out of eight tasks and higher success rates in six tasks. Furthermore, real-world experiments validate the practical feasibility of the learned policies, demonstrating the potential of "Next-Future" in solving complex robotic-arm tasks. Copyright © 2025, The Authors. All rights reserved.

关键词： Robotic arms

Enhancing Weakly Supervised Semantic Segmentation with Multi-modal Foundation Models: An End-to-End Approach

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ravanbakhsh, Elham Niu, Cheng Liang, Yongqing Ramanujam, J. Li, Xin Louisiana State University Baton RougeLA70803 United States Department of Computer Science & Engineering Texas A&M University College StationTX77843 United States Section of Visual Computing and Interactive Media Texas A&M University College StationTX77843 United States

Semantic segmentation is a core computer vision problem, but the high costs of data annotation have hindered its wide application. Weakly-Supervised Semantic Segmentation (WSSS) offers a cost-efficient workaround to extensive labeling in comparison to fully-supervised methods by using partial or incomplete labels. Existing WSSS methods have difficulties in learning the boundaries of objects leading to poor segmentation results. We propose a novel and effective framework that addresses these issues by leveraging visual foundation models inside the bounding box. Adopting a two-stage WSSS framework, our proposed network consists of a pseudo-label generation module and a segmentation module. The first stage leverages Segment Anything Model (SAM) to generate high-quality pseudo-labels. To alleviate the problem of delineating precise boundaries, we adopt SAM inside the bounding box with the help of another pre-trained foundation model (e.g., Grounding-DINO). Furthermore, we eliminate the necessity of using the supervision of image labels, by employing CLIP in classification. Then in the second stage, the generated high-quality pseudo-labels are used to train an off-the-shelf segmenter that achieves the state-of-the-art performance on PASCAL VOC 2012 and MS COCO 2014. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

SC3EF: A Joint Self-Correlation and Cross-Correspondence Estimation Framework for Visible and Thermal Image Registration

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Tong, Xi Luo, Xing Yang, Jiangxin Cao, Yanpeng Li, Xin State Key Laboratory of Fluid Power and Mechatronic Systems Laboratory of Advanced Manufacturing Technology of Zhejiang Province School of Mechanical Engineering Zhejiang University Hangzhou310027 China Section of Visual Computing and Creative Media School of Performance Visualization and Fine Arts Texas A&M University College StationTX77843 United States

Multispectral imaging plays a critical role in a range of intelligent transportation applications, including advanced driver assistance systems (ADAS), traffic monitoring, and night vision. However, accurate visible and thermal (RGB-T) image registration poses a significant challenge due to the considerable modality differences. In this paper, we present a novel joint Self-Correlation and Cross-Correspondence Estimation Framework (SC3EF), leveraging both local representative features and global contextual cues to effectively generate RGB-T correspondences. For this purpose, we design a convolution-transformer-based pipeline to extract local representative features and encode global correlations of intra-modality for inter-modality correspondence estimation between unaligned visible and thermal images. After merging the local and global correspondence estimation results, we further employ a hierarchical optical flow estimation decoder to progressively refine the estimated dense correspondence maps. Extensive experiments demonstrate the effectiveness of our proposed method, outperforming the current state-of-the-art (SOTA) methods on representative RGB-T datasets. Furthermore, it also shows competitive generalization capabilities across challenging scenarios, including large parallax, severe occlusions, adverse weather, and other cross-modal datasets (e.g., RGB-N and RGB-D). Copyright © 2025, The Authors. All rights reserved.

关键词： Advanced driver assistance systems

Deep Video Representation Learning: A Survey

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ravanbakhsh, Elham Liang, Yongqing Ramanujam, J. Li, Xin Division of Electrical & Computer Engineering Center for Computation & Technology Louisiana State University Baton RougeLA70803 United States Department of Computer Science and Engineering Texas A&M University College StationTX77843 United States Section of Visual Computing and Interactive Media Texas A&M University College StationTX77843 United States

This paper provides a review on representation learning for videos. We classify recent spatio-temporal feature learning methods for sequential visual data and compare their pros and cons for general video analysis. Building effective features for videos is a fundamental problem in computer vision tasks involving video analysis and understanding. Existing features can be generally categorized into spatial and temporal features. Their effectiveness under variations of illumination, occlusion, view and background are discussed. Finally, we discuss the remaining challenges in existing deep video representation learning studies. © 2024, CC BY-NC-ND.

关键词： Deep learning

Geodesign in the era of artificial intelligence

学校读者我要写书评

暂无评论

Frontiers of urban and rural planning 2025年第1期3卷 4页

作者： Xinyue Ye Tianchen Huang Yang Song Xin Li Galen Newman Zhongjie Lin Dayong Jason Wu Department of Landscape Architecture and Urban Planning & Center for Geospatial Sciences Applications and Technology Texas A&M University College Station TX 77840 USA. Section of Visual Computing and Computational Media School of Performance Texas A&M University Visualization and Fine Arts College Station TX 77840 USA. Weitzman School of Design University of Pennsylvania Philadelphia PA 19104 USA. Research & Implementation Texas A&M Transportation Institute Dallas TX 75251 USA.

This paper explores the evolution of Geodesign in addressing spatial and environmental challenges from its early foundations to the recent integration of artificial intelligence (AI). AI enhances existing Geodesign methods by automating spatial data analysis, improving land use classification, refining heat island effect assessment, optimizing energy use, facilitating green infrastructure planning, and generating design scenarios. Despite the transformative potential of AI in Geodesign, challenges related to data quality, model interpretability, and ethical concerns such as privacy and bias persist. This paper highlights case studies that demonstrate the application of AI in Geodesign, offering insights into its role in understanding existing systems and designing future changes. The paper concludes by advocating for the responsible and transparent integration of AI to ensure equitable and effective Geodesign outcomes.

关键词： Artificial Intelligence Ethics Geodesign Spatial planning