Semantic Occlusion Augmentation for Effective Human Pose Estimation


KIPS Transactions on Software and Data Engineering, Vol. 11, No. 12, pp. 517-524, Dec. 2022
https://doi.org/10.3745/KTSDE.2022.11.12.517,   PDF Download:  
Keywords: data augmentation, occlusion, Human Pose Estimation, Deep Learning
Abstract

Human pose estimation is a method of estimating a posture by extracting a human joint key point. When occlusion occurs, the joint key point extraction performance is lowered because the human joint is covered. The occlusion phenomenon is largely divided into three types of actions: self-contained, covered by other objects, and covered by background. In this paper, we propose an effective posture estimation method using a masking phenomenon enhancement technique. Although the posture estimation method has been continuously studied, research on the occlusion phenomenon of the posture estimation method is relatively insufficient. To solve this problem, the author proposes a data augmentation technique that intentionally masks human joints. The experimental results in this paper show that the intentional use of the blocking phenomenon enhancement technique is strong against the blocking phenomenon and the performance is increased.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
H. Bae, J. Kim, J. Lee, "Semantic Occlusion Augmentation for Effective Human Pose Estimation," KIPS Transactions on Software and Data Engineering, vol. 11, no. 12, pp. 517-524, 2022. DOI: https://doi.org/10.3745/KTSDE.2022.11.12.517.

[ACM Style]
Hyun-Jae Bae, Jin-Pyung Kim, and Jee-Hyong Lee. 2022. Semantic Occlusion Augmentation for Effective Human Pose Estimation. KIPS Transactions on Software and Data Engineering, 11, 12, (2022), 517-524. DOI: https://doi.org/10.3745/KTSDE.2022.11.12.517.