检索结果-南通市图书馆

FilterGNN:Image feature matching with cascaded outlier filters and linearattention

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第5期10卷 873-884页

作者： Jun-Xiong Cai Tai-Jiang Mu yu-kun lai Key Laboratory of Pervasive Computing Ministry of EducationDepartment of Computer Science and TechnologyTsinghua UniversityBeijing 100084China School of Computer Science and Informatics Cardiff UniversityWales CF244AGUK

The cross-view matching of local image features is a fundamental task in visual localization and 3D *** study proposes FilterGNN,a transformer-based graph neural network(GNN),aiming to improve the matching efficiency and accuracy of visual *** on high matching sparseness and coarse-to-fine covisible area detection,FilterGNN utilizes cascaded optimal graph-matching filter modules to dynamically reject outlier ***,we successfully adapted linear attention in FilterGNN with post-instance normalization support,which significantly reduces the complexity of complete graph learning from O(N2)to O(N).Experiments show that FilterGNN requires only 6%of the time cost and 33.3%of the memory cost compared with SuperGlue under a large-scale input size and achieves a competitive performance in various tasks,such as pose estimation,visual localization,and sparse 3D reconstruction.

关键词： image matching transformer linear attention visual localization sparse reconstruction

Benchmarking visual SLAM methods in mirror environments

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第2期10卷 215-241页

作者： Peter Herbert Jing Wu Ze Ji yu-kun lai School of Computer Science and Informatics Cardiff UniversityAbacws BuildingSenghennydd RdCardiff CF244AGUK School of Engineering Cardiff UniversityQueen’s BuildingsThe ParadeCardiff CF243AAUK

Visual simultaneous localisation and mapping(vSLAM)finds applications for indoor and outdoor navigation that routinely subjects it to visual complexities,particularly mirror *** effect of mirror presence(time visible and its average size in the frame)was hypothesised to impact localisation and mapping performance,with systems using direct techniques expected to perform ***,a dataset,MirrEnv,of image sequences recorded in mirror environments,was collected,and used to evaluate the performance of existing representative *** ORB-SLAM3 and BundleFusion appear to show moderate degradation of absolute trajectory error with increasing mirror duration,whilst the remaining results did not show significantly degraded localisation *** mesh maps generated proved to be very inaccurate,with real and virtual reflections colliding in the reconstructions.A discussion is given of the likely sources of error and robustness in mirror environments,outlining future directions for validating and improving vSLAM performance in the presence of planar *** MirrEnv dataset is available at https://***/10.17035/d.2023.0292477898.

关键词： visual simultaneous localisation and mapping(vSLAM) mirror localisation mapping reflection datase

A Revisit of Shape Editing Techniques:From the Geometric to the Neural Viewpoint

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2021年第3期36卷 520-554页

作者： yu-Jie yuan yu-kun lai Tong Wu Lin Gao Ligang Liu Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of SciencesBeijing 100190China University of Chinese Academy of Sciences Beijing 100049China School of Computer Science and Informatics Cardiff UniversityCardiff CF243AAU.K. School of Mathematical Sciences University of Science and Technology of ChinaHefei 230026China

3D shape editing is widely used in a range of applications such as movie production,computer games and computer aided *** is also a popular research topic in computer graphics and computer *** past decades,researchers have developed a series of editing methods to make the editing process faster,more robust,and more ***,the deformed shape is determined by the optimal transformation and weights for an energy *** increasing availability of 3D shapes on the Internet,data-driven methods were proposed to improve the editing *** recently as the deep neural networks became popular,many deep learning based editing methods have been developed in this field,which are naturally *** mainly survey recent research studies from the geometric viewpoint to those emerging neural deformation techniques and categorize them into organic shape editing methods and man-made model editing *** traditional methods and recent neural network based methods are reviewed.

关键词： mesh deformation man-made model editing deformation representation optimization deep learning

Saliency guided local and global descriptors for effective action recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2016年第1期2卷 97-106页

作者： Ashwan Abdulmunem yu-kun lai Xianfang Sun School of Computer Science and Informatics Cardiff University Department of Computer Science School of ScienceKerbala University

This paper presents a novel framework for human action recognition based on salient object detection and a new combination of local and global descriptors. We first detect salient objects in video frames and only extract features for such *** then use a simple strategy to identify and process only those video frames that contain salient objects. Processing salient objects instead of all frames not only makes the algorithm more efficient, but more importantly also suppresses the interference of background pixels. We combine this approach with a new combination of local and global descriptors, namely3D-SIFT and histograms of oriented optical flow(HOOF), respectively. The resulting saliency guided3D-SIFT–HOOF(SGSH) feature is used along with a multi-class support vector machine(SVM) classifier for human action recognition. Experiments conducted on the standard KTH and UCF-Sports action benchmarks show that our new method outperforms the competing state-of-the-art spatiotemporal feature-based human action recognition methods.

关键词： action recognition saliency detection local and global descriptors bag of visual words(Bo VWs) classification

ClusterSLAM:A SLAM backend for simultaneous rigid body clustering and motion estimation

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2021年第1期7卷 87-101页

作者： Jiahui Huang Sheng Yang Zishuo Zhao yu-kun lai Shi-Min Hu BNRist Department of Computer Science and TechnologyTsinghua UniversityBeijing 100084China Alibaba A.I.Labs Hangzhou 311121China School of Computer Science and Informatics Cardiff UniversityCardiffCF243AAUK

We present a practical backend for stereo visual SLAM which can simultaneously discover individual rigid bodies and compute their motions in dynamic *** recent factor graph based state optimization algorithms have shown their ability to robustly solve SLAM problems by treating dynamic objects as outliers,their dynamic motions are rarely *** this paper,we exploit the consensus of 3 D motions for landmarks extracted from the same rigid body for clustering,and to identify static and dynamic objects in a unified ***,our algorithm builds a noise-aware motion affinity matrix from landmarks,and uses agglomerative clustering to distinguish rigid *** decoupled factor graph optimization to revise their shapes and trajectories,we obtain an iterative scheme to update both cluster assignments and motion estimation *** on both synthetic scenes and KITTI demonstrate the capability of our approach,and further experiments considering online efficiency also show the effectiveness of our method for simultaneously tracking ego-motion and multiple objects.

关键词： dynamic SLAM motion segmentation scene perception

MILI:Multi-person inference from a low-resolution image

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Fundamental Research 2023年第3期3卷 434-441页

作者： kun Li yunke Liu yu-kun lai Jingyu Yang Tianjin University Tianjin 300350China CardiffUniversity CardiffCF244AGUnited Kingdom

Existing multi-person reconstruction methods require the human bodies in the input image to occupy a considerable portion of the ***,low-resolution human objects are ubiquitous due to trade-offbetween the field of view and target distance given a limited camera *** this paper,we propose an end-to-end multi-task framework for multi-person inference from a low-resolution image(MILI).To perceive more information from a low-resolution image,we use pair-wise images at high resolution and low resolution for training,and design a restoration network with a simple loss for better feature extraction from the low-resolution *** address the occlusion problem in multi-person scenes,we propose an occlusion-aware mask prediction network to estimate the mask of each person during 3D mesh *** results on both small-scale scenes and large-scale scenes demonstrate that our method outperforms the state-of-the-art methods both quantitatively and *** code is available at http://***/faculty/likun/projects/MILI.

关键词： Multi-person reconstruction Low-resolution human objects End-to-end Multi-task learning Occlusion-aware prediction

3D indoor scene modeling from RGB-D data:a survey

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2015年第4期1卷 267-278页

作者： Kang Chen yu-kun lai Shi-Min Hu Tsinghua University Cardiff University

3D scene modeling has long been a fundamental problem in computer graphics and computer vision. With the popularity of consumer-level RGB-D cameras,there is a growing interest in digitizing real-world indoor 3D scenes. However,modeling indoor3 D scenes remains a challenging problem because of the complex structure of interior objects and poor quality of RGB-D data acquired by consumer-level *** methods have been proposed to tackle these challenges. In this survey,we provide an overview of recent advances in indoor scene modeling techniques,as well as public datasets and code libraries which can facilitate experiments and evaluation.

关键词： RGB-D camera 3D indoor scenes geometric modeling semantic modeling survey

A Survey on Human Performance Capture and Animation

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2017年第3期32卷 536-554页

作者： Shihong Xia Lin Gao yu-kun lai Ming-Ze yuan Jinxiang Chai Advanced Computing Research Laboratory Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China School of Computer Science and Informatics Cardiff University Cardiff CF243AA U.K. University of Chinese Academy of Sciences Beijing 100049 China Department of Computer Science and Engineering Texas A&M University College Station Texas 77843-3112 U.S.A.

With the rapid development of computing technology, three-dimensional （3D） human body models and their dynamic motions are widely used in the digital entertainment industry. Human performance mainly involves human body shapes and motions. Key research problems in human performance animation include how to capture and analyze static geometric appearance and dynamic movement of human bodies, and how to simulate human body motions with physical effects. In this survey, according to the main research directions of human body performance capture and animation, we summarize recent advances in key research topics, namely human body surface reconstruction, motion capture and synthesis, as well as physics-based motion simulation, and further discuss future research problems and directions. We hope this will be helpful for readers to have a comprehensive understanding of human performance capture and animation.

关键词： human surface reconstruction body motion capture motion synthesis physics-based motion simulation

RecStitchNet:Learning to stitch images with rectangular boundaries

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第4期10卷 687-703页

作者： yun Zhang yu-kun lai Lang Nie Fang-Lue Zhang Lin Xu Key Lab of Film and TV Media Technology of Zhejiang Province College of Media EngineeringCommunication University of ZhejiangHangzhou 310018China School of Computer Science and Informatics Cardiff UniversityCardiff CF244AGUK Institute of Information Science Beijing Jiaotong UniversityBeijing 100091China School of Engineering and Computer Science Victoria University of WellingtonWellington 6012New Zealand STEM University of South AustraliaAdelaide 5095Australia

Irregular boundaries in image stitching naturally occur due to freely moving *** deal with this problem,existing methods focus on optimizing mesh warping to make boundaries regular using the traditional explicit ***,previous methods always depend on hand-crafted features(e.g.,keypoints and line segments).Thus,failures often happen in overlapping regions without distinctive *** this paper,we address this problem by proposing RecStitchNet,a reasonable and effective network for image stitching with rectangular *** that both stitching and imposing rectangularity are non-trivial tasks in the learning-based framework,we propose a three-step progressive learning based strategy,which not only simplifies this task,but gradually achieves a good balance between stitching and imposing *** the first step,we perform initial stitching by a pre-trained state-of-the-art image stitching model,to produce initially warped stitching results without considering the boundary ***,we use a regression network with a comprehensive objective regarding mesh,perception,and shape to further encourage the stitched meshes to have rectangular boundaries with high content ***,we propose an unsupervised instance-wise optimization strategy to refine the stitched meshes iteratively,which can effectively improve the stitching results in terms of feature alignment,as well as boundary and structure *** to the lack of stitching datasets and the difficulty of label generation,we propose to generate a stitching dataset with rectangular stitched images as pseudo-ground-truth labels,and the performance upper bound induced from the it can be broken by our unsupervised *** and quantitative results and evaluations demonstrate the advantages of our method over the state-of-the-art.

关键词： image stitching boundaries convolutional neural network