Self-Supervised Neural Aggregation Networks for Human Parsing

Jian Zhao, Jianshu Li, Xuecheng Nie, Fang Zhao, Yunpeng Chen, Zhecan Wang, Jiashi Feng, Shuicheng Yan

Abstract

In this paper, we present a Self-Supervised Neural Aggregation Network (SS-NAN) for human parsing. SS-NAN adaptively learns to aggregate the multi-scale features at each pixel "address". In order to further improve the feature discriminative capacity, a self-supervised joint loss is adopted as an auxiliary learning strategy, which imposes human joint structures into parsing results without resorting to extra supervision. The proposed SS-NAN is end-to-end trainable. SS-NAN can be integrated into any advanced neural networks to help aggregate features regarding the importance at different positions and scales and incorporate rich high-level knowledge regarding human joint structures from a global perspective, which in turn improve the parsing results. Comprehensive evaluations on the recent Look into Person (LIP) and the PASCAL-Person-Part benchmark datasets demonstrate the significant superiority of our method over other state-of-the-arts.
BibTeX
@inproceedings{zhao2017self,
  author = {Jian Zhao and Jianshu Li and Xuecheng Nie and Fang Zhao and Yunpeng Chen and Zhecan Wang and Jiashi Feng and Shuicheng YAN},
  title = {{Self-Supervised Neural Aggregation Networks for Human Parsing}},
  booktitle = {Workshops @ CVPR},
  year = {2017},
  doi = {10.1109/cvprw.2017.204},
}

Loading related content...

Figure