TY - GEN
T1 - PCNET
T2 - 28th IEEE International Conference on Image Processing, ICIP 2021
AU - Wang, Jianyuan
AU - You, Meiyue
AU - Leng, Biao
AU - Jiang, Ming
AU - Song, Guanglu
N1 - Publisher Copyright:
© 2021 IEEE
PY - 2021
Y1 - 2021
N2 - Person re-identification has a wide range of applications, and many state-of-the-art methods are proposed to solve the problem under specific scenarios. However, it is still a challenging issue because of the large variance in practical applications, such as pose variations, misalignment, and image noises. In this paper, Parallelly Conquer Net (PCNet) is proposed to deal with large variance in a parallel manner. PCNet consists of three module: Pose Adaptation Module (PAM), Global Alignment Module (GAM), and Pixel-Wised Attention Module (PWAM). Each module is designed to deal with a sub-variance independently. Furthermore, the generated features are aggregated by parallel branches to utilize complementary information among them. Extensive experiments on three benchmarks (Market-1501, DukeMTMC-reID, and CUHK03) demonstrate the effectiveness of the method. The results show that PCNet can significantly improve the performance of person re-identification.
AB - Person re-identification has a wide range of applications, and many state-of-the-art methods are proposed to solve the problem under specific scenarios. However, it is still a challenging issue because of the large variance in practical applications, such as pose variations, misalignment, and image noises. In this paper, Parallelly Conquer Net (PCNet) is proposed to deal with large variance in a parallel manner. PCNet consists of three module: Pose Adaptation Module (PAM), Global Alignment Module (GAM), and Pixel-Wised Attention Module (PWAM). Each module is designed to deal with a sub-variance independently. Furthermore, the generated features are aggregated by parallel branches to utilize complementary information among them. Extensive experiments on three benchmarks (Market-1501, DukeMTMC-reID, and CUHK03) demonstrate the effectiveness of the method. The results show that PCNet can significantly improve the performance of person re-identification.
KW - Alignment
KW - Parallelly conquer net
KW - Person re-identification
KW - Pixel attention
KW - Pose adaptation
UR - https://www.scopus.com/pages/publications/85125562995
U2 - 10.1109/ICIP42928.2021.9506010
DO - 10.1109/ICIP42928.2021.9506010
M3 - 会议稿件
AN - SCOPUS:85125562995
T3 - Proceedings - International Conference on Image Processing, ICIP
SP - 2368
EP - 2372
BT - 2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings
PB - IEEE Computer Society
Y2 - 19 September 2021 through 22 September 2021
ER -