Unsupervised Person Re Identification On 5

Metrics

MAP

Rank-1

Rank-10

Rank-5

Results

Performance results of various models on this benchmark

					Paper Title
TMGF	76.8	86.7	94.1	92.9	Transformer Based Multi-Grained Features for Unsupervised Person Re-Identification
VAL-PAT	74.9	86.1	-	-	Learning Transferable Pedestrian Representation from Multimodal Information Supervision
HHCL(ResNet50 w/o RK)	73.3	85.1	94.6	92.4	Hard-sample Guided Hybrid Contrast Learning for Unsupervised Person Re-Identification
ICE	69.9	83.3	94.1	91.5	ICE: Inter-instance Contrastive Encoding for Unsupervised Person Re-identification
Group Sampling	69.1	82.7	93.5	91.1	Rethinking Sampling Strategies for Unsupervised Person Re-identification
CACL	69.6	82.6	93.8	91.2	Cluster-guided Asymmetric Contrastive Learning for Unsupervised Person Re-Identification
SpCL	65.3	81.2	92.2	90.3	Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID
CAP	67.3	81.1	91.8	89.3	Camera-aware Proxies for Unsupervised Person Re-Identification
Self-Similarity Grouping (one shot)	55.9	72.4	87.7	84	Self-similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-identification
ECN	40.4	63.3	80.4	75.8	Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identification
SPGAN+LMP	26.2	46.4	68.0	62.3	Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification
PUL	16.4	30.0	48.5	43.4	Unsupervised Person Re-identification: Clustering and Fine-tuning
PTGAN	-	27.4	50.7	-	Person Transfer GAN to Bridge Domain Gap for Person Re-Identification

0 of 13 row(s) selected.

Unsupervised Person Re Identification On 5 | SOTA | HyperAI