fast-reid/MODEL_ZOO.md

FastReID Model Zoo and Baselines

Introduction

BoT:

Bag of Tricks and A Strong Baseline for Deep Person Re-identification. CVPRW2019, Oral.

AGW:

This is a re-implementation of ReID-Survey with a Powerful AGW Baseline

MGN:

SBS:

stronger baseline on top of BoT:

Bag of Freebies(BoF):

1. Circle loss
2. Freeze backbone training
3. Cutout data augmentation & Auto Augmentation
4. Cosine annealing learning rate decay
5. Soft margin triplet loss

Bag of Specials(BoS):

1. Non-local block
2. GeM pooling

Market1501 Baselines

BoT:

Method	Pretrained	Rank@1	mAP	mINP
BoT(R50)	ImageNet	94.4%	86.1%	59.4%
BoT(R50-ibn)	ImageNet	94.9%	87.6%	64.1%
BoT(S50)	ImageNet	95.1%	88.5%	66.0%
BoT(R101-ibn)	ImageNet	95.4%	88.9%	67.4%

AGW:

Method	Pretrained	Rank@1	mAP	mINP
AGW(R50)	ImageNet	95.3%	88.2%	66.3%
AGW(R50-ibn)	ImageNet	95.1%	88.7%	67.1%
AGW(S50)	ImageNet	94.7%	87.1%	62.2%
AGW(R101-ibn)	ImageNet	95.5%	89.5%	69.5%

SBS:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50)	ImageNet	95.4%	88.2%	64.8%
SBS(R50-ibn)	ImageNet	95.7%	89.3%	67.5%
SBS(S50)	ImageNet	95.0%	87.0%	60.6%
SBS(R101-ibn)	ImageNet	96.3%	90.3%	70.0%

MGN:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50-ibn)	ImageNet	95.8%	89.7%	67.0%

DukeMTMC Baseline

BoT:

Method	Pretrained	Rank@1	mAP	mINP
BoT(R50)	ImageNet	87.1%	76.9%	41.6%
BoT(R50-ibn)	ImageNet	89.6%	79.1%	44.4%
BoT(S50)	ImageNet	87.8%	77.7%	39.6%
BoT(R101-ibn)	ImageNet	91.1%	81.3%	47.7%

AGW:

Method	Pretrained	Rank@1	mAP	mINP
AGW(R50)	ImageNet	89.0%	79.9%	46.3%
AGW(R50-ibn)	ImageNet	89.8%	80.7%	47.7%
AGW(S50)	ImageNet	89.9%	79.7%	44.2%
AGW(R101-ibn)	ImageNet	91.4%	82.1%	50.2%

SBS:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50)	ImageNet	89.6%	79.8%	44.6%
SBS(R50-ibn)	ImageNet	91.3%	81.6%	47.6%
SBS(S50)	ImageNet	90.5%	79.1%	42.7%
SBS(R101-ibn)	ImageNet	92.4%	83.2%	49.7%

MGN:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50-ibn)	ImageNet	91.6%	82.1%	46.7%

MSMT17 Baseline

BoT:

Method	Pretrained	Rank@1	mAP	mINP
BoT(R50)	ImageNet	72.3%	48.3%	9.7%
BoT(R50-ibn)	ImageNet	77.0%	54.4%	12.5%
BoT(S50)	ImageNet	80.4%	59.2%	15.9%
BoT(R101-ibn)	ImageNet	79.0%	57.5%	14.6%

AGW:

Method	Pretrained	Rank@1	mAP	mINP
AGW(R50)	ImageNet	76.7%	53.6%	12.2%
AGW(R50-ibn)	ImageNet	79.3%	57.5%	14.3%
AGW(S50)	ImageNet	77.3%	54.7%	12.6%
AGW(R101-ibn)	ImageNet	80.8%	60.2%	16.5%

SBS:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50)	ImageNet	83.3%	59.9%	14.6%
SBS(R50-ibn)	ImageNet	84.0%	61.2%	15.5%
SBS(S50)	ImageNet	82.6%	58.2%	13.2%
SBS(R101-ibn)	ImageNet	85.1%	63.3%	16.6%

MGN:

Method	Pretrained	Rank@1	mAP	mINP
SBS(R50)	ImageNet	82.9%	61.2%	14.9%
SBS(R50-ibn)	ImageNet	85.1%	65.4%	18.4%
SBS(R101-ibn)	ImageNet	96.3%	90.3%	70.0%

VeRi Baseline

SBS:

Method	Pretrained	Rank@1	mAP	mINP
BoT(R50-ibn)	ImageNet	97.0%	81.3%	44.9%

VehicleID Baseline

BoT: Method: BoT(R50-ibn+gem pooling+weighted triplet+soft margin) Test protocol: 10-fold cross-validation

Testset size	Pretrained	Rank@1	Rank@5
Small(800)	ImageNet	86.6%	97.9%
Medium(1600)	ImageNet	82.9%	96.0%
Large(2400)	ImageNet	80.6%	93.9%

VERI-Wild Baseline

BoT: Method: BoT(R50-ibn+gem pooling+weighted triplet+soft margin)

Testset size	Pretrained	Rank@1	mAP	mINP
Small(3000)	ImageNet	96.4%	87.7%	69.2%
Medium(5000)	ImageNet	95.1%	83.5%	61.2%
Large(10000)	ImageNet	92.5%	77.3%	49.8%