Person Re Identification On Market 1501

评估指标

Rank-1

mAP

评测结果

各个模型在此基准测试上的表现结果

模型名称	Rank-1	mAP	Paper Title	Repository
CAL	95.5	89.5	Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification	-
Auto-ReID(RK)	95.4	94.2	Auto-ReID: Searching for a Part-aware ConvNet for Person Re-Identification	-
OSNet	94.8	84.9	Omni-Scale Feature Learning for Person Re-Identification	-
Top-DB-Net + RK	95.5	94.1	Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification	-
BPBreID	95.7	89.4	Body Part-Based Representation Learning for Occluded Person Re-Identification	-
DAAF-BoT	95.1	87.9	Deep Attention Aware Feature Learning for Person Re-Identification	-
S-CNN	65.88	39.55	Gated Siamese Convolutional Neural Network Architecture for Human Re-Identification	-
MGN + CircleLoss(ours)	96.1	87.4	Circle Loss: A Unified Perspective of Pair Similarity Optimization	-
A3M	86.54	68.97	Attribute-Aware Attention Model for Fine-grained Representation Learning	-
P2-Net (triplet loss)	95.2	85.6	Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification	-
PyrNet (single-shot)	93.6	81.7	Aggregating Deep Pyramidal Representations for Person Re-Idenfitication
SPreID [kalayeh2018human]	93.6	83.3	Human Semantic Parsing for Person Re-identification	-
CBN+BoT*	94.3	83.6	Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization	-
CLIP-ReID Baseline +UFFM+AMC	96.1	92.0	Enhancing person re-identification via Uncertainty Feature Fusion Method and Auto-weighted Measure Combination	-
IDE*	85.66	65.87	Camera Style Adaptation for Person Re-identification	-
LuNet	81.38	60.71	In Defense of the Triplet Loss for Person Re-Identification	-
Dual Cluster Contrastive	95.4	89.2	Dual Cluster Contrastive learning for Object Re-Identification	-
MSINet (2.3M w/o RK)	95.3	89.6	MSINet: Twins Contrastive Search of Multi-Scale Interaction for Object ReID	-
DiP (without RK)	95.8	90.8	DiP: Learning Discriminative Implicit Parts for Person Re-Identification	-
KPR	95.9	89.6	Keypoint Promptable Re-Identification	-

0 of 135 row(s) selected.