Person Re Identification On Market 1501 C

Metrics

Rank-1

mAP

mINP

Results

Performance results of various models on this benchmark

				Paper Title
TransReID	53.19	27.38	1.98	TransReID: Transformer-based Object Re-Identification
CaceNet	42.92	18.24	0.67	Devil in the Details: Towards Accurate Single and Multiple Human Parsing
PLR-OS	37.56	14.23	0.48	Learning Diverse Features with Part-Level Resolution for Person Re-Identification
RRID	36.57	14.23	0.48	Relation Network for Person Re-identification
Pyramid	35.72	12.75	0.36	Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training
PCB	34.93	12.72	0.41	Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)
SBS (ResNet-50)	34.13	11.54	0.29	FastReID: A Pytorch Toolbox for General Instance Re-identification
BDB	33.79	10.95	0.32	Batch DropBlock Network for Person Re-identification and Beyond
MHN	33.29	10.69	0.38	Mixed High-Order Attention Network for Person Re-Identification
AGW (ResNet-50)	31.90	12.13	0.35	Deep Learning for Person Re-identification: A Survey and Outlook
DG-Net	31.75	9.96	0.35	Joint Discriminative and Generative Learning for Person Re-identification
VPM	31.17	10.15	0.31	Perceive Where to Focus: Learning Visibility-aware Part-level Features for Partial Person Re-identification
Aligned++	31.00	10.95	0.32	AlignedReID: Surpassing Human-Level Performance in Person Re-Identification
OS-Net	30.94	10.37	0.23	Omni-Scale Feature Learning for Person Re-Identification
ABD-Net	29.65	9.81	0.26	ABD-Net: Attentive but Diverse Person Re-Identification
MGN	29.56	9.72	0.29	Learning Discriminative Features with Multiple Granularities for Person Re-Identification
F-LGPR	29.35	9.08	0.23	A Person Re-identification Data Augmentation Method with Adversarial Defense Effect
TDB	28.56	8.90	0.20	Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
LGPR	27.72	8.26	0.24	A Person Re-identification Data Augmentation Method with Adversarial Defense Effect
BoT (ResNet-50)	27.05	8.42	0.20	Bag of Tricks and A Strong Baseline for Deep Person Re-identification

0 of 22 row(s) selected.

Command Palette

Person Re Identification On Market 1501 C

Metrics

Results