.png){kind=icon}
.png){kind=icon}
Return
Keypoint based weakly supervised human parsing
.png){kind=icon}
OA
AI
DOI：10.1016/j.imavis.2019.08.005.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Fully convolutional networks (FCN) have achieved great success in human parsing in recent years. In conventional human parsing tasks, pixel-level labeling is required for guiding the training, which usually involve enormous human labeling efforts. To ease the labeling efforts, we propose a novel weakly supervised human parsing method which only requires simple object keypoint annotations for learning. We develop an itertive learning method to generate pseudo part segmentation masks from keypoint labels. With these pseud masks, we train a FCN network to output pixel-level human parsing predictions. Furthermore, we develop correlation network to perform joint prediction of part and object segmentation masks and improve the segmentation performance. The experiment results show that our weakly supervised method is able to achies very competitive human parsing results. Despite that our method only uses simple keypoint annotatior for learning, we are able to achieve comparable performance with fully supervised methods which use the expensive pixel-level annotations. (C) 2019 Elsevier B.V. All rights reservesed
Keywords:
Human parsing
Weakly supervise
Iterative refinement
Keypoint
Skeleton
Correlation network
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.2
Papers:
4.0K
Citations:
6.7K