search

SHI-Labs / Self-Similarity-Grouping

Self-similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-identification (ICCV 2019, Oral)
187 stars 43 forks source link

features about rerank #14

Open zengkaiwei opened 5 years ago

zengkaiwei commented 5 years ago

Usually rerank is used in one dataset,but in your rerank.py you rerank source feature with target feature? Why? Or my understanding is fault? At this moment, what's the mean of your e_dist and r_dist? Is it still distence between each sample in target datasets?

xiaobaoli15 commented 4 years ago

I have also the same question.

tiancity-NJU commented 4 years ago

same question too

OasisYang commented 4 years ago

Sorry for the late reply, I was really busy. Your understanding is right, we use re-rank between source feature and target feature. The motivation is to obtain a better re-rank result by the knowledge of source dataset. (since the model can achieve pretty good performance on source dataset). e_dist is the euclidean distance among samples in a target dataset. r_dist is the distance after re-ranking. Hope this can solve your problem.

JingweiZhang12 commented 4 years ago

Did you try to use the original dist rather than source dist? What's the performance compared to the current source dist? image