578 Hits in 5.8 sec

A Data-driven Approach for Human Pose Tracking Based on Spatio-temporal Pictorial Structure [article]

Soumitra Samanta, Bhabatosh Chanda
2016 arXiv   pre-print
In this paper, we present a data-driven approach for human pose tracking in video data.  ...  We formulate the human pose tracking problem as a discrete optimization problem based on spatio-temporal pictorial structure model and solve this problem in a greedy framework very efficiently.  ...  et al. [56] present a tracking method based on spatio-temporal context learning.  ... 
arXiv:1608.00199v1 fatcat:ynm65dtvpfgoxcbs46uawbqk74

Automatic Pose Tracking and Motion Transfer to Arbitrary 3D Characters [chapter]

Ju Shen, Jianjun Yang
2015 Lecture Notes in Computer Science  
In order to reduce the ambiguity of the estimated 3D pose, a modified spatio-temporal constraint based algorithm is used for articulated gesture estimation across frames while maintaining temporal coherence  ...  Our approach demonstrates promising performance on par with state-of-theart techniques.  ...  Pose Estimation from Monocular Pictorial Structures To achieve fully automatic 2D human pose tracking from a video, we start with the state-of-the-art mixture of parts detectors by representing the human  ... 
doi:10.1007/978-3-319-21978-3_56 fatcat:mcofjzkw7nhkrpcp2xauh7ydi4

Interactive activity recognition using pose-based spatio–temporal relation features and four-level Pachinko Allocation Model

Thien Huynh-The, Ba-Vui Le, Sungyoung Lee, Yongik Yoon
2016 Information Sciences  
We propose a novel approach that permits to deeply understand complex personperson activities based on the knowledge coming from human pose analysis.  ...  We validate our interaction recognition method on two practical data sets, the BIT-Interaction data set and the UT-Interaction data set.  ...  Furthermore, a shape-based kernel for upper-body pose similarity and a leave-one-out loss function were developed in the learning phase. Building on a successful pictorial structures model, Tian et al  ... 
doi:10.1016/j.ins.2016.06.016 fatcat:xaesiv7qivdy7la53h2qmukk54

Special issue: Behaviours in video

Huiyu Zhou, Yuan Yuan, Eliza Y Du, Pingkun Yan
2013 Neurocomputing  
Acknowledgement The Guest Editors would like to thank all the authors for submitting their high quality manuscripts to this special issue, and all the reviewers for providing their timely review comments  ...  Vera Kamphuis for encouraging and supporting this special issue. Guest Editors Huiyu Zhou, Ph.D. Yuan Yuan, Ph.D. Yingzi Du, Ph.D. Pingkun Yan, Ph.D.  ...  Bouziane et al. presented a unified framework for human behavior recognition, using Markov spatio-temporal random walks on graph.  ... 
doi:10.1016/j.neucom.2012.04.018 fatcat:l7vob67pybev3dwscg6p5lqfhq

Monocular human pose tracking using multi frame part dynamics

Vivek Kumar Singh, Ramakant Nevatia
2009 2009 Workshop on Motion and Video Computing (WMVC)  
We present a monte-carlo approximation of the body dynamics using spatio-temporal distributions over part tracks.  ...  To obtain tracks that favor kinematically feasible body poses, we propose a novel "kinematically constrained" particle filtering approach which results in more accurate pose tracking than other stochastic  ...  We present an approximation of the spatio-temporal volume of human pose dynamics using part tracks and proposed a principled approach to efficiently sample kinematically consistent part tracks.  ... 
doi:10.1109/wmvc.2009.5399247 fatcat:3sgwaous5zgqdfit4kzjrq77gq

Athlete Pose Estimation from Monocular TV Sports Footage

Mykyta Fastovets, Jean-Yves Guillemaut, Adrian Hilton
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
Belief-propagation over a spatio-temporal graph of candidate body part hypotheses is used to estimate a temporally consistent pose between key-frame constraints.  ...  We propose an interactive modelbased generative approach for estimating the human pose in 2D from uncalibrated monocular video in unconstrained sports TV footage without any prior learning on motion captured  ...  Acknowledgements: The authors would like to thank BBC R&D, Graham Thomas and Robert Dawes for supporting this research and providing data.  ... 
doi:10.1109/cvprw.2013.152 dblp:conf/cvpr/FastovetsGH13 fatcat:yjt27mouezek7fpqk4iafdjb3q

Author Index

2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Detecting and Parsing Architecture at City Scale from Range Data Adaptive Pose Priors for Pictorial Structures Talking Pictures: Temporal Grouping and Dialog-Supervised Person Recognition Workshop  ...  Discovering Spatio-Temporal Dependencies in Dynamic Scenes Tracking the Invisible: Learning Where the Object Might be A Hough Transform-Based Voting Framework for Action Recognition Finding Nemo: Deformable  ... 
doi:10.1109/cvpr.2010.5539913 fatcat:y6m5knstrzfyfin6jzusc42p54

Space-Time Representation of People Based on 3D Skeletal Data: A Review [article]

Fei Han, Brian Reily, William Hoff, Hao Zhang
2017 arXiv   pre-print
Spatiotemporal human representation based on 3D visual perception data is a rapidly growing research area.  ...  Based on the information sources, these representations can be broadly categorized into two groups based on RGB-D information or 3D skeleton data.  ...  Black, Pose-conditioned joint angle limits for 3D hu- 3D pictorial structures for multiple human pose estimation, in: IEEE man pose reconstruction, in: IEEE Conference on Computer Vision  ... 
arXiv:1601.01006v3 fatcat:zgpmw3xqunajtctu4e7minjyx4

Action is in the Eye of the Beholder: Eye-gaze Driven Model for Spatio-Temporal Action Localization

Nataliya Shapovalova, Michalis Raptis, Leonid Sigal, Greg Mori
2013 Neural Information Processing Systems  
We propose a weakly-supervised structured learning approach for recognition and spatio-temporal localization of actions in video.  ...  As part of the proposed approach, we develop a generalization of the Max-Path search algorithm which allows us to efficiently search over a structured space of multiple spatio-temporal paths while also  ...  In [7] an extension to multiple sub-volumes that model parts of the action is proposed and amounts to a spatio-temporal part-based (pictorial structure) model.  ... 
dblp:conf/nips/ShapovalovaRSM13 fatcat:eoczf3q77vbplnyrsew7xaxt5y

Custom Pictorial Structures for Re-identification

Dong Seon Cheng, Marco Cristani, Michele Stoppa, Loris Bazzani, Vittorio Murino
2011 Procedings of the British Machine Vision Conference 2011  
We propose a novel methodology for re-identification, based on Pictorial Structures (PS).  ...  It is based on the statistical learning of pixel attributes collected through spatio-temporal reasoning.  ...  In this paper, we present a novel methodology for human re-id, based on Pictorial Structures (PS) for human body pose estimation.  ... 
doi:10.5244/c.25.68 dblp:conf/bmvc/ChengCSBM11 fatcat:hmdnnkst4jhv7e4tybzdttd2rm

A survey of advances in vision-based human motion capture and analysis

Thomas B. Moeslund, Adrian Hilton, Volker Krüger
2006 Computer Vision and Image Understanding  
A number of significant research advances are identified together with novel methodologies for automatic initialization, tracking, pose estimation and movement recognition.  ...  This survey reviews recent trends in video based human capture and analysis, as well as discussing open problems for future research to achieve automatic visual analysis of human movement.  ...  Moeslund and Granum [205,211] apply a data driven sequential Monte Carlo approach to pose estimation of a human arm.  ... 
doi:10.1016/j.cviu.2006.08.002 fatcat:7vsbfnczrzgsbdbbmlzvfprh54

A Survey on Visual Surveillance of Object Motion and Behaviors

W. Hu, T. Tan, L. Wang, S. Maybank
2004 IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews)  
Visual surveillance in dynamic scenes, especially for humans and vehicles, is currently one of the most active research topics in computer vision.  ...  and description of behaviors, human identification, and fusion of data from multiple cameras.  ...  Xu from the NLPR for their valuable suggestions and assistance in preparing this paper.  ... 
doi:10.1109/tsmcc.2004.829274 fatcat:cozxn2ogtrew3pybyuxcrj2rhi

Recent Advances in Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspective [article]

Wu Liu, Qian Bao, Yu Sun, Tao Mei
2021 arXiv   pre-print
We believe this survey will provide the readers with a deep and insightful understanding of monocular human pose estimation.  ...  Estimation of the human pose from a monocular camera has been an emerging research topic in the computer vision community with many applications.  ...  Moreover, instead of the manually defined body structure relation, they propose a data-driven approach to group related parts based on the amount of information they shared.  ... 
arXiv:2104.11536v1 fatcat:tdag2jq2vjdrjekwukm5nu7l6a

Video based activity recognition in trauma resuscitation

Ishani Chakraborty, Ahmed Elgammal, Randall S. Burd
2013 2013 10th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG)  
In this multi-agent, multitask setting, we represent procedures as high-level concepts composed of low-level features based on the patient's pose, scene dynamics, clinician motions and device locations  ...  At runtime, a Markov Network is dynamically constructed representing hypothesized procedures, spatio-temporal relationships and attribute probabilities.  ...  In this method, a weak human body detector identifies possible body locations followed by a stronger pictorial structure based model that enforces kinematic constraints on body parts.  ... 
doi:10.1109/fg.2013.6553758 dblp:conf/fgr/ChakrabortyEB13 fatcat:2ug24mlhkbggddsrbago2sqfwi

Recent Advances of Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspective

Wu Liu, Tao Mei
2022 ACM Computing Surveys  
Estimation of the human pose from a monocular camera has been an emerging research topic in the computer vision community with many applications.  ...  Then we summarize the mainstream and milestone approaches for these human body presentations since the year 2014 under unified frameworks.  ...  Instead of using the shared features for all body parts, Tang et al. [192] propose a data-driven approach to group related parts based on the amount of information they shared.  ... 
doi:10.1145/3524497 fatcat:4pbvntngrnfp7lqhcpjmy7p2fq
« Previous Showing results 1 — 15 out of 578 results