ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Adaptive Directional Walks For Pose Estimation From Single Body Depths
Cited 2 time in scopus Share share facebook twitter linkedin kakaostory
Authors
Jaehwan Kim, Junsuk Lee
Issue Date
2020-06
Citation
International Conference on Multimedia and Expo (ICME) 2020, pp.1-6
Publisher
IEEE
Language
English
Type
Conference Paper
DOI
https://dx.doi.org/10.1109/ICME46284.2020.9102922
Abstract
In this paper, we introduce a novel body pose estimation method based on single depth images with our proposed random forest classifier, whereby it is possible to estimate the positions of joints directly with significant accuracy. We train randomized classification trees based on ajoint entropy objective function combined with the geodesic distances and directional vectors simultaneously, to estimate the probability distribution for the label of the directional vector towards a particular bodyjoint by considering the geodesic structural information. At a test step, an arbitrary point moves as much as the magnitude of the probability predicted adaptively by following the predefined kinematic graph, which is referred to as adaptive directional walks. Numerical and visual experiments with real datasets confirm the usefulness of the proposed method.
KSP Keywords
Classification tree, Estimation method, Geodesic distance, Pose estimation, Probability distribution, Structural information, Visual experiments, objective function, random forest classifier, single depth images