Filters








10,282 Hits in 4.2 sec

Combining Local Appearance and Holistic View: Dual-Source Deep Neural Networks for Human Pose Estimation [article]

Xiaochuan Fan, Kang Zheng, Yuewei Lin, Song Wang
2015 arXiv   pre-print
We propose a new learning-based method for estimating 2D human pose from a single image, using Dual-Source Deep Convolutional Neural Networks (DS-CNN).  ...  Recently, many methods have been developed to estimate human pose by using pose priors that are estimated from physiologically inspired graphical models or learned from a holistic perspective.  ...  Related Work Part-based models for human pose estimation.  ... 
arXiv:1504.07159v1 fatcat:bd2bpodhsvgj3jwhrc46ltsyuq

Combining local appearance and holistic view: Dual-Source Deep Neural Networks for human pose estimation

Xiaochuan Fan, Kang Zheng, Yuewei Lin, Song Wang
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose a new learning-based method for estimating 2D human pose from a single image, using Dual-Source Deep Convolutional Neural Networks (DS-CNN).  ...  Recently, many methods have been developed to estimate human pose by using pose priors that are estimated from physiologically inspired graphical models or learned from a holistic perspective.  ...  Related Work Part-based models for human pose estimation.  ... 
doi:10.1109/cvpr.2015.7298740 dblp:conf/cvpr/FanZLW15 fatcat:a6dyha2cmfcmdl5n4j7oajtgxi

Predicting 3D People from 2D Pictures [chapter]

Leonid Sigal, Michael J. Black
2006 Lecture Notes in Computer Science  
The resulting 3D estimates are sufficiently accurate to serve as proposals for the Bayesian inference of 3D human motion over time.  ...  First we infer a learned view-based 2D body model from a single image using non-parametric belief propagation.  ...  Likelihoods The likelihood model for an individual limb is built to account for possible occlusions between body parts for a given view-based 2D model.  ... 
doi:10.1007/11789239_19 fatcat:ilvj2swqffak3a3hyyxt3sqwla

Detailed Human Shape and Pose from Images

Alexandru O. Balan, Leonid Sigal, Michael J. Black, James E. Davis, Horst W. Haussecker
2007 2007 IEEE Conference on Computer Vision and Pattern Recognition  
Much of the research on video-based human motion capture assumes the body shape is known a priori and is represented coarsely (e.g. using cylinders or superquadrics to model limbs).  ...  These body models stand in sharp contrast to the richly detailed 3D body models used by the graphics community. Here we propose a method for recovering such models directly from images.  ...  The relative shapes of these "metaballs" were defined a priori and were then scaled for each limb based on an estimated length and width parameter for that limb.  ... 
doi:10.1109/cvpr.2007.383340 dblp:conf/cvpr/BalanSBDH07 fatcat:m6j7sl6mvjdzfbzgybix4qap7q

Single upper limb pose estimation method based on improved stacked hourglass network [article]

Gang Peng, Yuezhi Zheng, Jianfeng Li, Jin Yang, Zhonghua Deng
2020 arXiv   pre-print
For use in human-machine cooperative operations, this paper proposes a single-person upper limb pose estimation method based on an end-to-end approach for accurate and real-time limb pose estimation.  ...  Experiments showed that the developed single-person upper limb skeleton key point detection model achieves high accuracy and that the pose estimation method based on the end-to-end approach provides high  ...  The single upper limb pose estimation based on the cascade method uses a human detector and single upper limb skeleton key point detection model.  ... 
arXiv:2004.07456v1 fatcat:joybz45q3ncgtkgegvtskl4pme

Computer Vision to the Rescue: Infant Postural Symmetry Estimation from Incongruent Annotations [article]

Xiaofei Huang, Michael Wan, Lingfei Luan, Bethany Tunik, Sarah Ostadabbas
2022 arXiv   pre-print
In this paper, we develop a computer vision based infant symmetry assessment system, leveraging 3D human pose estimation for infants.  ...  To rectify this, we develop a Bayesian estimator of the ground truth derived from a probabilistic graphical model of fallible human raters.  ...  Specifically, our strategy is as follows: • To remedy the lackluster human rater reliability, we employ a probabilistic graphical model of the human raters as fallible assessors, and compute a Bayesian  ... 
arXiv:2207.09352v1 fatcat:ltcgq3t2yfgyrcy6y3raml7py4

Real-time Human Pose Estimation from Body-scanned Point Clouds

Jilliam María Díaz Barros, Frederic Garcia, Désiré Sidibé
2015 Proceedings of the 10th International Conference on Computer Vision Theory and Applications  
This paper presents a novel approach to estimate the human pose from a body-scanned point cloud.  ...  To do so, a predefined skeleton model is first initialized according to both the skeleton base point and its torso limb obtained by Principal Component Analysis (PCA).  ...  ., 2009) , where the authors estimate the 2-D human pose in a video sequence using a predefined human skeleton model to fit the silhouette of a body shape.  ... 
doi:10.5220/0005309005530560 dblp:conf/visapp/BarrosGS15 fatcat:ul537zmelvb4rfkcftdhd2dduq

Using Rao-Blackwellised Particle Filter Track 3D Arm Motion Based On Hierarchical Limb Model

XueSong Yu, JiaFeng Liu, XiangLong Tang, JianHua Huang
2008 Zenodo  
For improving the efficiency of human 3D tracking, we present an algorithm to track 3D Arm Motion. First, the Hierarchy Limb Model (HLM) is proposed based on the human 3D skeleton model.  ...  Finally, Rao-Blackwellised Particle Filter is used to estimate the 3D arm motion. The result of experiment shows that our algorithm can advance the computation efficiency.  ...  Recently, the structure graphical model [9] has been used to facilitate the estimation of human 3D motion. Based on the structure graphical model, Wei [13] proposed a decentralized framework.  ... 
doi:10.5281/zenodo.1083524 fatcat:aoaaagd2cjhirc3l4pdya47w5a

Combination of Annealing Particle Filter and Belief Propagation for 3D Upper Body Tracking

Ilaria Renna, Ryad Chellali, Catherine Achard
2012 Applied Bionics and Biomechanics  
3D upper body pose estimation is a topic greatly studied by the computer vision society because it is useful in a great number of applications, mainly for human robots interactions including communications  ...  To tackle this problem, we propose a new approach that combines several annealing particle filters defined independently for each limb and belief propagation method to add geometrical constraints between  ...  Human pose estimation can be obtained in a tracking context, or directly, by analysing video images.  ... 
doi:10.1155/2012/178981 fatcat:3hl2h7sn4beebgjcjd4metvgry

Human Pose Tracking Using Multi-level Structured Models [chapter]

Mun Wai Lee, Ram Nevatia
2006 Lecture Notes in Computer Science  
We proposed a three-stage approach with a multi-level state representation that enables a hierarchical estimation of 3D body poses. At the first stage, humans are tracked as blobs.  ...  In the second stage, parts such as face, shoulders and limbs are estimated and estimates are combined by grid-based belief propagation to infer 2D joint positions.  ...  We use a graphical model to represent the human body.  ... 
doi:10.1007/11744078_29 fatcat:swcdjya6mbeyjaqxu3f625heai

Human Pose Tracking in Monocular Sequence Using Multilevel Structured Models

Mun Wai Lee, R. Nevatia
2009 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Index Terms-3D human pose estimation, multiple-human tracking, belief propagation, data driven Markov chain Monte Carlo, generative models.  ...  We proposed a three-stage approach with multilevel state representation that enables a hierarchical estimation of 3D body poses.  ...  We use a graphical model to represent the human body.  ... 
doi:10.1109/tpami.2008.35 pmid:19029544 fatcat:tprs2vm2i5bbtoqd24hp6nxb6i

Characterizing Multidimensional Capacitive Servoing for Physical Human-Robot Interaction [article]

Zackory Erickson, Henry M. Clever, Vamsee Gangaram, Eliot Xing, Greg Turk, C. Karen Liu, Charles C. Kemp
2021 arXiv   pre-print
Capacitive servoing then uses these human pose estimates from a data-driven pose estimator within a feedback control loop in order to maneuver the robot's end effector around the surface of a human limb  ...  human limb.  ...  We feed these measurements to a trained pose estimation model which provides a relative 4D pose estimate for a nearby human limb.  ... 
arXiv:2105.11582v2 fatcat:fcerrwyogbh3ja6fqxwfrwy57e

A Semi-Supervised Data Augmentation Approach using 3D Graphical Engines [article]

Shuangjun Liu, Sarah Ostadabbas
2018 arXiv   pre-print
A state-of-the-art human pose estimation deep learning model then was trained from scratch using our ScanAva dataset and could achieve the pose estimation accuracy of 91.2% at PCK0.5 criteria after applying  ...  This presents a fundamental problem for applications with limited, expensive, or private data (i.e. small data), such as human pose and behavior estimation/tracking which could be highly personalized.  ...  Pose Estimation Performance of Trained DNN Models To evaluate the quality of the synthesized datasets, we employed a state-of-theart DNN-based 2D human pose estimation algorithm, a stacked hourglass model  ... 
arXiv:1808.02595v2 fatcat:s7mxadnvuvgavdftbrbwimfdde

Survey of Motion Tracking Methods Based on Inertial Sensors: A Focus on Upper Limb Human Motion

Alessandro Filippeschi, Norbert Schmitz, Markus Miezal, Gabriele Bleser, Emanuele Ruffaldi, Didier Stricker
2017 Sensors  
After a survey on IMU-based human tracking, five techniques for motion reconstruction were selected and compared to reconstruct a human arm motion.  ...  Results show that all but one of the selected models perform similarly (about 35 mm average position estimation error).  ...  Hol et al. propose an alternative approach to IHMT for pose estimation of one limb based on UWB [101] .  ... 
doi:10.3390/s17061257 pmid:28587178 pmcid:PMC5492902 fatcat:xvdft2sf3re77k5l3ebkjgcvra

A Framework for Human Pose Estimation in Videos [article]

Dong Zhang, Mubarak Shah
2016 arXiv   pre-print
In this paper, we present a method to estimate a sequence of human poses in unconstrained videos.  ...  We aim to demonstrate that by using temporal information, the human pose estimation results can be improved over image based pose estimation methods.  ...  TREE-BASED OPTIMIZATION FOR HUMAN POSE ESTIMATION IN VIDEOS We formulate the video based human pose estimation problem into a unified tree-based optimization framework, which can be solved efficiently  ... 
arXiv:1604.07788v1 fatcat:rqhyy4ko7rhmvby2sirttcfmtq
« Previous Showing results 1 — 15 out of 10,282 results