Bag of Tricks and A Strong Baseline for Deep Person Re-identification

被引：971

作者：

Luo, Hao ^{[1
]}

Gu, Youzhi ^{[1
]}

Liao, Xingyu ^{[2
]}

Lai, Shenqi ^{[3
]}

Jiang, Wei ^{[1
]}

机构：

[1] Zhejiang Univ, Hangzhou, Zhejiang, Peoples R China

[2] Chinese Acad Sci, Beijing, Peoples R China

[3] Xi An Jiao Tong Univ, Xian, Peoples R China

来源：

2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2019) | 2019年

基金：

中国国家自然科学基金;

关键词：

D O I：

10.1109/CVPRW.2019.00190

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper explores a simple and efficient baseline for person re-identification (ReID). Person re-identification (ReID) with deep neural networks has made progress and achieved high performance in recent years. However, many state-of-the-arts methods design complex network structure and concatenate multi-branch features. In the literature, some effective training tricks are briefly appeared in several papers or source codes. This paper will collect and evaluate these effective training tricks in person ReID. By combining these tricks together, the model achieves 94.5% rank-I and 85.9% mAP on Market1501 with only using global features. Our codes and models are available at https://github.com/michuanhaohao/reid-strong-baseline

引用

页码：1487 / 1495

页数：9

共 28 条

[1]

[Anonymous], 2016, CoRR abs/1512.00567, DOI DOI 10.1109/CVPR.2016.308

[2]

[Anonymous], 2018, ARXIV181006996

[3]

Dai Zuozhuo, 2018, ARXIV PREPRINT ARXIV

[4]

Fan X., 2019, J VISUAL COMMUNICATI

[5]

He K., 2016, CVPR, DOI [10.1109/CVPR.2016.90, DOI 10.1109/CVPR.2016.90]

[6]

Hermans Alexander, 2017, ARXIV

[7] Human Semantic Parsing for Person Re-identification [J].

Kalayeh, Mahdi M. ;

Basaran, Emrah ;

Gokmen, Muhittin ;

Kamasak, Mustafa E. ;

Shah, Mubarak .

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :1062-1071

[8]

Li DP, 2015, PROC CVPR IEEE, P213, DOI 10.1109/CVPR.2015.7298617

[9] Harmonious Attention Network for Person Re-Identification [J].

Li, Wei ;

Zhu, Xiatian ;

Gong, Shaogang .

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :2285-2294

[10]

Qi L., 2018, ARXIV180403864

← 1 2 3 →