For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Occluded Appearance Modeling with Sample Weighting for Human Pose Estimation
Yuki KAWANA Norimichi UKITA
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2017/10/01
Online ISSN: 1745-1361
Type of Manuscript: PAPER
Category: Image Recognition, Computer Vision
human pose estimation, pictorial structure models, occlusion,
Full Text: PDF>>
This paper proposes a method for human pose estimation in still images. The proposed method achieves occlusion-aware appearance modeling. Appearance modeling with less accurate appearance data is problematic because it adversely affects the entire training process. The proposed method evaluates the effectiveness of mitigating the influence of occluded body parts in training sample images. In order to improve occlusion evaluation by a discriminatively-trained model, occlusion images are synthesized and employed with non-occlusion images for discriminative modeling. The score of this discriminative model is used for weighting each sample in the training process. Experimental results demonstrate that our approach improves the performance of human pose estimation in contrast to base models.