End-to-End Thorough Body Perception for Person Search

Authors

  • Kun Tian Horizon Robotics
  • Houjing Huang Chinese Academy of Sciences
  • Yun Ye Horizon Robotics
  • Shiyu Li Horizon Robotics
  • Jinbin Lin Horizon Robotics
  • Guan Huang Horizon Robotics

DOI:

https://doi.org/10.1609/aaai.v34i07.6886

Abstract

In this paper, we propose an improved end-to-end multi-branch person search network to jointly optimize person detection, re-identification, instance segmentation, and keypoint detection. First, we build a better and faster base model to extract non-highly correlated feature expression; Second, a foreground feature enhance module is used to alleviate undesirable background noise in person feature maps; Third, we design an algorithm to learn the part-aligned representation for person search. Extensive experiments with ablation analysis show the effectiveness of our proposed end-to-end multi-task model, and we demonstrate its superiority over the state-of-the-art methods on two benchmark datasets including CUHK-SYSU and PRW.

Downloads

Published

2020-04-03

How to Cite

Tian, K., Huang, H., Ye, Y., Li, S., Lin, J., & Huang, G. (2020). End-to-End Thorough Body Perception for Person Search. Proceedings of the AAAI Conference on Artificial Intelligence, 34(07), 12079-12086. https://doi.org/10.1609/aaai.v34i07.6886

Issue

Section

AAAI Technical Track: Vision