Person Re Identification On Market 1501 C

Rank-1

mAP

mINP

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	Rank-1	mAP	mINP
omni-scale-feature-learning-for-person-re	30.94	10.37	0.23
alignedreid-surpassing-human-level	31.00	10.95	0.32
a-person-re-identification-data-augmentation	27.72	8.26	0.24
devil-in-the-details-towards-accurate-single	42.92	18.24	0.67
benchmarks-for-corruption-invariant-person-re	-	-	-
mixed-high-order-attention-network-for-person	33.29	10.69	0.38
abd-net-attentive-but-diverse-person-re	29.65	9.81	0.26
batch-feature-erasing-for-person-re	33.79	10.95	0.32
top-db-net-top-dropblock-for-activation	28.56	8.90	0.20
joint-discriminative-and-generative-learning	31.75	9.96	0.35
relation-network-for-person-re-identification	36.57	14.23	0.48
fastreid-a-pytorch-toolbox-for-real-world	34.13	11.54	0.29
transreid-transformer-based-object-re	53.19	27.38	1.98
bags-of-tricks-and-a-strong-baseline-for-deep	27.05	8.42	0.20
learning-discriminative-features-with	29.56	9.72	0.29
beyond-part-models-person-retrieval-with	34.93	12.72	0.41
deep-learning-for-person-re-identification-a	31.90	12.13	0.35
a-coarse-to-fine-pyramidal-model-for-person	35.72	12.75	0.36
unsupervised-pre-training-for-person-re	-	-	-
learning-diverse-features-with-part-level	37.56	14.23	0.48
a-person-re-identification-data-augmentation	29.35	9.08	0.23
perceive-where-to-focus-learning-visibility	31.17	10.15	0.31