This paper tackles the problem of reconstructing 3D human poses from given 2D landmarks, which is still an ill-posed problem. The existing works have successfully applied Active Shape Model approach to estimate 3D human poses, but the execution time is quite high. In this paper, we propose a speed-up method by separating data into subspaces to reduce the execution time of existing methods in two steps: (i) Predicting the subspace that the need-estimated 3D shape could belong to. (ii) Estimating 3D shape from given 2D landmarks and predefined basis shapes of this subspace. Compare to existing works; our approach shows a significant reduction in computational time.
CITATION STYLE
Hoang, V. T., & Jo, K. H. (2018). Speed-Up 3D Human Pose Estimation Task Using Sub-spacing Approach. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10752 LNAI, pp. 553–562). Springer Verlag. https://doi.org/10.1007/978-3-319-75420-8_52
Mendeley helps you to discover research relevant for your work.