9,770 Hits in 4.9 sec

Geometric constraints on 2D action models for tracking human body

Alexei Gritai, Arslan Basharat, Mubarak Shah
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
We propose a 2D model-based approach for tracking human body parts during articulated motion.  ...  The novelty of our approach lies in the geometric formulation of human actions and the use of geometric constraints for body joints estimation.  ...  The formulation of the geometric constraints on the geome-try of human action is novel.  ... 
doi:10.1109/icpr.2008.4761437 dblp:conf/icpr/GritaiBS08 fatcat:dyltp2dg2zhobpnbsv4cg6p5jm

Action recognition in cluttered dynamic scenes using Pose-Specific Part Models

Vivek Kumar Singh, Ram Nevatia
2011 2011 International Conference on Computer Vision  
We adopt a Joint Tracking and Recognition approach, which track the actor pose by sampling from 3D action models.  ...  We present an approach to recognizing single actor human actions in complex backgrounds.  ...  We address these issues by a more accurate localization of the human pose using a 2D part model with kinematic constraints.  ... 
doi:10.1109/iccv.2011.6126232 dblp:conf/iccv/SinghN11 fatcat:okygguqznvgzhaw7parp5s4uau

Geometric Pose Affordance: 3D Human Pose with Scene Constraints [article]

Zhe Wang, Liyan Chen, Shaurya Rathore, Daeyun Shin, Charless Fowlkes
2021 arXiv   pre-print
that encourages learned models to favor geometrically consistent pose estimates.  ...  To inject prior knowledge of scene constraints into existing frameworks for pose estimation from images, we introduce a novel, view-based representation of scene geometry, a multi-layer depth map, which  ...  DOPE: Distillation Of Part Experts for whole-body 3D pose estimation in the wild, in: ECCV. Xiao, B., Wu, H., Wei, Y., 2018. Simple baselines for human pose estimation and tracking, in: ECCV.  ... 
arXiv:1905.07718v2 fatcat:rr6fj6jx5fbvzawdk7ssou6x6e

Human Motion Analysis: A Review

J.K. Aggarwal, Q. Cai
1999 Computer Vision and Image Understanding  
We focus on three major areas related to interpreting human motion: 1) motion analysis involving human body parts, 2) tracking of human motion wing single or multiple cameras, and 8) recognizing human  ...  Ilfacking human motion wing a single or multiple cameras focuses on higher-level processing, in which moving humans are tracked without identifying specific parts of the body structure.  ...  Motion on Human Body PartsFigure 2: Past research on motion analysis of human body parts.Figure 3 Past research on tracking of human motion without using body parts.  ... 
doi:10.1006/cviu.1998.0744 fatcat:uazu7o5qerfzfnnokfpks6k3ju

Leaving Flatland: Advances in 3D behavioral measurement [article]

Jesse D. Marshall, Tianqing Li, Joshua H. Wu, Timothy W. Dunn
2021 arXiv   pre-print
Existing 3D measurement techniques draw on specialized hardware, such as motion capture or depth cameras, as well as deep multi-view and monocular computer vision.  ...  Continued advances at the intersection of deep learning and computer vision will facilitate 3D tracking across more anatomical features, with less training data, in additional species, and within more  ...  While 2D measurements can be used for action recognition, both [92, 96, 97] supervised [98] [97] and unsupervised [35, 39, 48] 2D recognition approaches depend on camera viewing angle.  ... 
arXiv:2112.01987v1 fatcat:z5dqj47s3fbt3d4t75bnsp2l6q

Human Motion Tracking with Less Constraint of Initial Posture from a Single RGB-D Sensor

Chen Liu, Anna Wang, Chunguang Bu, Wenhui Wang, Haijing Sun
2021 Sensors  
High-quality and complete human motion 4D reconstruction is of great significance for immersive VR and even human operation.  ...  However, it has inevitable self-scanning constraints, and tracking under monocular settings also has strict restrictions.  ...  Subsequently, we tracked the human body motion and combined the priors and depth to refine the geometric details of the model.  ... 
doi:10.3390/s21093029 pmid:33925847 fatcat:wgzlbubwhnczphvpsqc3rc2gzm

View-Invariant Action Recognition [chapter]

Yogesh Singh Rawat, Shruti Vyas
2020 Computer Vision  
The varying pattern of spatio-temporal appearance generated by human action is key for identifying the performed action.  ...  The research in view-invariant action recognition addresses this problem and focuses on recognizing human actions from unseen viewpoints.  ...  The human body has certain anthropometric proportion, and there exist some geometric constraints between multiple anatomical landmarks such as body joints.  ... 
doi:10.1007/978-3-030-03243-2_878-1 fatcat:up2g5kc6h5borf56tvg3vj4ljm

Predicting Intentions of Pedestrians from 2D Skeletal Pose Sequences with a Representation-Focused Multi-Branch Deep Learning Network

Joseph Gesnouin, Steve Pechberti, Guillaume Bresson, Bogdan Stanciulescu, Fabien Moutarde
2020 Algorithms  
Understanding the behaviors and intentions of humans is still one of the main challenges for vehicle autonomy.  ...  Hence, we propose SPI-Net (Skeleton-based Pedestrian Intention network): a representation-focused multi-branch network combining features from 2D pedestrian body poses for the prediction of pedestrians  ...  The authors would like to thank Anna Venia, Marie Morel and Philéas Samir for proofreading the article.  ... 
doi:10.3390/a13120331 fatcat:e6kox7drkjesbkomh7fdedhvvu

Interactive generation of human animation with deformable motion models

Jianyuan Min, Yen-Lin Chen, Jinxiang Chai
2009 ACM Transactions on Graphics  
Figure 1: Intuitive interactive motion generation with deformable motion models: (a) direct manipulation interfaces with point dragging (red point) and fixed handles (green point); (b) pen-based sketching  ...  However, the statistical models used for human motion tracking have limitations that are similar to the models used for human motion synthesis.  ...  For example, it can be used to generate 3D animation sequences from 2D animation sequences, reconstruct 3D human motion from a small set of 2D trajectories tracked from video, or interpolate intermediate  ... 
doi:10.1145/1640443.1640452 fatcat:nbfqkp67b5grbiujw77p7bdxka

3D Body Reconstruction for Immersive Interaction [chapter]

Isaac Cohen, Mun Wai Lee
2002 Lecture Notes in Computer Science  
The 3D reconstruction is based on the representation of body parts using Generalized Cylinders providing an estimation of the 3D shape of the human body.  ...  the observed shape and the learned descriptions for inferring the articulated body model.  ...  Acknowledgment This research was supported by a grant from the Institute for Creative Technologies (  ... 
doi:10.1007/3-540-36138-3_10 fatcat:r2nqll72n5e33l66wknigqdkei

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  
Recent research has addressed reliable tracking and pose estimation in natural scenes. Progress has also been made towards automatic understanding of human actions and behavior.  ...  A number of significant research advances are identified together with novel methodologies for automatic initialization, tracking, pose estimation and movement recognition.  ...  Results are presented for 2D tracking of upper-body and walking motions in cluttered scenes.  ... 
doi:10.1016/j.cviu.2006.08.002 fatcat:7vsbfnczrzgsbdbbmlzvfprh54

Human Motion Capture Using a Drone

Xiaowei Zhou, Sikang Liu, Georgios Pavlakos, Vijay Kumar, Kostas Daniilidis
2018 2018 IEEE International Conference on Robotics and Automation (ICRA)  
In this work we introduce a drone-based system for 3D human MoCap.  ...  We argue that, besides the capability of tracking a moving subject, a flying drone also provides fast varying viewpoints, which is beneficial for motion reconstruction.  ...  First, obtaining clean 2D keypoint tracks from a video is difficult especially for fast moving and deformable objects like human body.  ... 
doi:10.1109/icra.2018.8462830 dblp:conf/icra/ZhouLP0D18 fatcat:llkywycf3jha5nhznhmq2oerba

Human Motion Capture Using a Drone [article]

Xiaowei Zhou, Sikang Liu, Georgios Pavlakos, Vijay Kumar, Kostas Daniilidis
2018 arXiv   pre-print
In this work we introduce a drone-based system for 3D human MoCap.  ...  We argue that, besides the capability of tracking a moving subject, a flying drone also provides fast varying viewpoints, which is beneficial for motion reconstruction.  ...  First, obtaining clean 2D keypoint tracks from a video is difficult especially for fast moving and deformable objects like human body.  ... 
arXiv:1804.06112v1 fatcat:uqlkuaibbjatraoxpnyvkce2we

Multi-View Matching (MVM): Facilitating Multi-Person 3D Pose Estimation Learning with Action-Frozen People Video [article]

Yeji Shen, C.-C. Jay Kuo
2020 arXiv   pre-print
The core technology of MVM lies in effective alignment of 2D poses obtained from multiple views of a static scene that has a strong geometric constraint.  ...  Our objective is to maximize mutual consistency of 2D poses estimated in multiple frames, where geometric constraints as well as appearance similarities are taken into account simultaneously.  ...  The parametric-body-model-based representation offers richer information of the human body. Early work used the SCAPE model [14] that fits a body model to annotated 2D keypoints in the image.  ... 
arXiv:2004.05275v1 fatcat:5xpqhxrm3bgynlqs54otj7hemm

3D Human Tracking in a Top View Using Depth Information Recorded by the Xtion Pro-Live Camera [chapter]

Cyrille Migniot, Fakhreddine Ababsa
2013 Lecture Notes in Computer Science  
The model is composed of two parts: a 2D model providing the human localization and a 3D model providing its pose.  ...  Human body poses could be estimated through model fitting using dense correspondences between depth data and an articulated human model.  ...  Introduction In recent years, there is a body of research on the problem of human parts detection, pose estimation and tracking from 3D data.  ... 
doi:10.1007/978-3-642-41939-3_59 fatcat:a2ygm7ygjrbszor46iv4d7a3je
« Previous Showing results 1 — 15 out of 9,770 results