Structured Domain Adaptation with Online Relation Regularization for Unsupervised Person Re-ID

14 Mar 2020 Yixiao Ge Feng Zhu Rui Zhao Hongsheng Li

Unsupervised domain adaptation (UDA) aims at adapting the model trained on a labeled source-domain dataset to an unlabeled target-domain dataset. The task of UDA on open-set person re-identification (re-ID) is even more challenging as the identities (classes) do not overlap between the two domains... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Unsupervised Domain Adaptation Duke to Market SDA mAP 70.0 # 4
rank-1 86.9 # 4
rank-5 94.4 # 4
rank-10 96.3 # 5
Unsupervised Domain Adaptation Duke to MSMT SDA mAP 25.6 # 3
rank-1 54.4 # 2
rank-5 66.4 # 2
rank-10 71.3 # 2
Unsupervised Domain Adaptation Market to Duke SDA mAP 61.4 # 4
rank-1 76.5 # 4
rank-5 86.6 # 4
rank-10 89.7 # 3
Unsupervised Domain Adaptation Market to MSMT SDA mAP 23.2 # 3
rank-1 49.5 # 3
rank-5 62.2 # 4
rank-10 67.7 # 4

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet