更新
2020年11月15日:添加了新的论文复现
- Pyramid [CVPR2019] Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training
目录
项目地址:https://github.com/nickhuang1996/HJL-re-id
一、简介
这是由博主自己完成的行人重识别代码库,包含了博主自己研究的MDRS模型。
该项目包含对日志记录、损失监测和可视化Rank图像的充分支持。
项目中的各个模型都是Pytorch实现的, 包含了PCB,MGN,PGFA, Pyramid和HOReID等顶会论文模型。
二、实现的Re-ID模型
- PCB [ECCV2018] Beyond Part Models: Person Retrieval with Refined Part Pooling (and A Strong Convolutional Baseline)
- MGN [ACM Multimedia 2018] Learning Discriminative Features with Multiple Granularities for Person Re-Identification
- PGFA [ICCV2019] Pose-Guided Feature Alignment for Occluded Person Re-Identification
- Pyramid [CVPR2019] Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training
- HOReID [CVPR2020] High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification
三、MDRS
- 名称: Joint multi-scale discrimination and region segmentation for person re-ID
- 期刊: Pattern Recognition Letters
- Volume 138, October 2020, Pages 540-547
- JCR: Q2
- web of science
框架
四、遮挡的行人重识别
博主提供了 PGFAReIDTrainer.py,PyramidalReIDTrainer.py和 HOReIDTrainer.py 来实现模型PGFA和模型HOReID的训练和测试
文章来源: nickhuang1996.blog.csdn.net,作者:悲恋花丶无心之人,版权归原作者所有,如需转载,请联系作者。
原文链接:nickhuang1996.blog.csdn.net/article/details/108699607