SHI-Labs/Self-Similarity-Grouping
Self-similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-identification (ICCV 2019, Oral)
Python
Issues
- 4
How to do PK sampler to ensure the calculating of three triplet losses when the labels pf one image are different?
#23 opened by wangyujie413 - 4
- 1
dataset写的比较难懂
#31 opened by guanyonglai - 0
License question
#38 opened by ragondyn - 1
TypeError: Can't instantiate abstract class Euclidean with abstract methods get_metric, score_pairs
#33 opened by zongzhesun - 13
Market2Duke results
#2 opened by jh97321 - 0
ValueError: => No checkpoint found
#37 opened by beiyangxiaolaodi - 0
where is the script "source_train.py"?
#29 opened by shuxjweb - 0
SSG+&&SSG++
#36 opened by talentCODE - 1
- 0
- 0
where is source_train.py
#34 opened by zzw-zwzhang - 0
- 0
The generation of the pseudo labels.
#30 opened by ZacharyWang-007 - 4
- 5
- 0
- 2
why reranking so time-cosuming?
#4 opened by luissen - 4
TypeError: Can't instantiate abstract class Euclidean with abstract methods get_metric, score_pairs
#10 opened by geyutang - 1
- 0
Seems a typo in the paper
#24 opened by SunskyF - 5
some problems in source_train.py
#1 opened by BravoLu - 4
features about rerank
#14 opened by zengkaiwei - 0
something confused me in JointTrainer2
#22 opened by zhufq00 - 0
something confused me in JointTrainer2
#21 opened by zhufq00 - 1
- 1
why use the same compute_dist way in both selftraining.py and semitraining.py?
#17 opened by NothingToSay99 - 2
About the source_train.py
#3 opened by luissen - 2
- 1
Error on source_train.py
#15 opened by mahfujur1 - 0
parameter about dce_loss
#13 opened by zengkaiwei - 0
- 1
So using re_ranking when pseudo label assignment is not using the re_ranking method?
#7 opened by luissen - 1
DukeMTMC-reID dataset download is broken, can you share this dataset download link?
#9 opened by geyutang - 1
some error of eug.py
#8 opened by zengkaiwei - 0
- 1