Nonparametric density estimation for human pose tracking

Authors:
Thomas Brox;Bodo Rosenhahn;Uwe G. Kersting;Daniel Cremers
Affiliations:
CVPR Group, University of Bonn, Bonn, Germany;MPI for Computer Science, Saarbrücken, Germany;Department of Sport and Exercise Science, The University of Auckland, New Zealand;CVPR Group, University of Bonn, Bonn, Germany
Venue:
DAGM'06 Proceedings of the 28th conference on Pattern Recognition
Year:
2006

Citing 11
Cited 7

A Mathematical Introduction to Robotic Manipulation

A Mathematical Introduction to Robotic Manipulation
Fast Pose Estimation with Parameter-Sensitive Hashing

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Unsupervised Non-parametric Region Segmentation Using Level Sets

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Twist Based Acquisition and Tracking of Animal and Human Kinematics

International Journal of Computer Vision
Generative modeling for continuous non-linearly embedded visual inference

ICML '04 Proceedings of the twenty-first international conference on Machine learning
Bayesian Body Localization Using Mixture of Nonlinear Shape Models

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 - Volume 01
Recovering 3D Human Pose from Monocular Images

IEEE Transactions on Pattern Analysis and Machine Intelligence
Kernel Density Estimation and Intrinsic Alignment for Shape Priors in Level Set Segmentation

International Journal of Computer Vision
Representation and matching of articulated shapes

CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Three-dimensional shape knowledge for joint image segmentation and pose estimation

PR'05 Proceedings of the 27th DAGM conference on Pattern Recognition
A nonparametric statistical method for image segmentation using information theory and curve evolution

IEEE Transactions on Image Processing

A system for articulated tracking incorporating a clothing model

Machine Vision and Applications
Staying Well Grounded in Markerless Motion Capture

Proceedings of the 30th DAGM symposium on Pattern Recognition
Optimization and Filtering for Human Motion Capture

International Journal of Computer Vision
Clustered stochastic optimization for object recognition and pose estimation

Proceedings of the 29th DAGM conference on Pattern recognition
Nonparametric density estimation with adaptive, anisotropic kernels for human motion tracking

Proceedings of the 2nd conference on Human motion: understanding, modeling, capture and animation
Learning for multi-view 3d tracking in the context of particle filters

ISVC'06 Proceedings of the Second international conference on Advances in Visual Computing - Volume Part II
Analyzing and evaluating markerless motion tracking using inertial sensors

ECCV'10 Proceedings of the 11th European conference on Trends and Topics in Computer Vision - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

The present paper considers the supplement of prior knowledge about joint angle configurations in the scope of 3-D human pose tracking. Training samples obtained from an industrial marker based tracking system are used for a nonparametric Parzen density estimation in the 12-dimensional joint configuration space. These learned probability densities constrain the image-driven joint angle estimates by drawing solutions towards familiar configurations. This prevents the method from producing unrealistic pose estimates due to unreliable image cues. Experiments on sequences with a human leg model reveal a considerably increased robustness, particularly in the presence of disturbed images and occlusions.