Person_reID_pytorch
from Resnet-50->Part-based Convolutional Baseline
It is consistent with the new baseline result in several top-conference works, e.g., Beyond Part Models: Person Retrieval with Refined Part Pooling(ECCV18) and Camera Style Adaptation for Person Re-identification(CVPR18). And arrived Rank@1=93.24%, mAP=80.78% only with softmax loss and Part-based Convolutional Baseline(PCB) method!
1.Random Erasing Augmentation has been added
2.Last Stridehas been adjusted 1\
文章解读
CSDN博客：小白入门计算机视觉系列——ReID(一)：什么是ReID？如何做ReID？ReID数据集？ReID评测指标？\