List your company
ResearchFormerly WizWand

Image Classification on ANIMAL-10N (test)

89.38AccuracyProposed Method
79.000881.695484.3987.0846Mar 13, 2021Jan 22, 2022Dec 4, 2022Oct 16, 2023Aug 26, 2024Jul 8, 2025May 20, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Proposed MethodModel Variant=Modified Model, Test-Time Augmentation=true2022.1089.38
Jigsaw-ViTBackbone=DeiT-Small/16, Number of Parameters=22M, FLOPs=4.6G2022.0789
MixFixBackbone=PreResNet-18, Zero-shot=false2024.0888.85
Ours2024.0888.85
S3Backbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0788.5
Robust LR2021.1288.5
SSR+2021.1188.5
SSR+Backbone=PreResNet-18, Zero-shot=false2024.0888.5
SSR+2024.0888.5
Proposed MethodModel Variant=Modified Model, Test-Time Augmentation=false2022.1088.48
GAMRBackbone=VGG-19N2026.0588.1
DeiT-Small/16Backbone=DeiT-Small/16, Number of Parameters=22M, FLOPs=4.6G2022.0787.2
LongReMixBackbone=VGG-19N2026.0587.2
DICSBackbone=VGG-19N2026.0587.1
Proposed MethodModel Variant=Normal Model, Test-Time Augmentation=true2022.1086.98
CTRRBackbone=ResNet-18, Pre-training=Random initialization2022.0386.71
PSSCLBackbone=VGG-19N2026.0586.2
Proposed MethodModel Variant=Normal Model, Test-Time Augmentation=false2022.1085.84
VRI (+)Backbone=VGG19, Metadata availability=None, Adaptive sample strategy=true2026.0385.8
BRBackbone=VGG-19N2026.0585.8
Co-teachingBackbone=ResNet-18, Pre-training=Random initialization2022.0385.73
OT-FilterBackbone=VGG-19N2026.0585.5
CMW-Net-SL2022.0284.7
CMW-Net-SL2022.0284.7
InstanceGMBackbone=VGG [52]2022.0984.6
InstanceGM2024.0884.6
Chen et al (2022)Backbone=VGG19, Metadata availability=None2026.0384.5
DivideMixBackbone=VGG-19N2026.0584.5
GCEBackbone=ResNet-18, Pre-training=Random initialization2022.0384.42
HRP (Ours)Backbone=VGG-19N2026.0584.3
APLBackbone=ResNet-18, Pre-training=Random initialization2022.0384.25
Englesson (2021)Backbone=VGG19, Metadata availability=None2026.0384.2
Nested + Co-teaching2021.0484.1
Nested-CE2022.0984.1
NCT2022.1084.1
NCTBackbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0784.1
NCT2021.1284.1
NCT2021.1184.1
NCT2024.0884.1
Chen et al (2021)Backbone=VGG19, Metadata availability=None2026.0384.1
Co-teaching+NestedBackbone=VGG-19N2026.0584.1
SLBackbone=ResNet-18, Pre-training=Random initialization2022.0383.92
ForwardBackbone=ResNet-18, Pre-training=Random initialization2022.0383.67
PLCbackbone=VGG-19 with batch normalization, optimizer=SGD, epochs=1002021.0383.4
PLC2021.0483.4
PLC2022.0983.4
PLC2022.1083.4
PLCBackbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0783.4
PLC2021.1283.4
PLC2022.0283.4
PLC2022.0283.4
PLC2021.1183.4
PLC2024.0883.4
Zhang et al (2021b)Backbone=VGG19, Metadata availability=None2026.0383.4
PLCBackbone=VGG-19N2026.0583.4
CEBackbone=ResNet-18, Pre-training=Random initialization2022.0383.18
SLNBackbone=ResNet-18, Pre-training=Random initialization2022.0383.17
MixUpBackbone=VGG-19N2026.0582.7
DALBackbone=ResNet-182026.0582.7
SELFIEbackbone=VGG-19 with batch normalization, optimizer=SGD, epochs=1002021.0381.8
SELFIE2021.0481.8
SELFIE2022.0981.8
SELFIE2022.1081.8
SELFIEBackbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0781.8
SELFIE2021.1281.8
SELFIE2022.0281.8
SELFIE2022.0281.8
SELFIE2021.1181.8
SELFIE2024.0881.8
Song et al (2019)Backbone=VGG19, Metadata availability=None2026.0381.8
SELFIEBackbone=VGG-19N2026.0581.8
DivideMixBackbone=VGG-19N2026.0581.5
GCEBackbone=VGG-19N2026.0581.5
VRI (-)Backbone=VGG19, Metadata availability=None, Adaptive sample strategy=false2026.0381.4
RGC2026.0581.4
L2BBackbone=VGG-19N2026.0581.4
Dropout2021.0481.3
Nested2021.0481.3
Nested-Dropout2022.0981.3
CE+Dropout2022.0981.3
DropoutBackbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0781.3
CMW-Net2022.0280.9
CMW-Net2022.0280.9
MW-Net2022.0280.7
MW-Net2022.0280.7
ActiveBias2022.0280.5
ActiveBias2022.0280.5
Co-teaching2022.0280.2
Co-teaching2022.0280.2
HamBRBackbone=VGG-19N2026.0579.7
Standardbackbone=VGG-19 with batch normalization, optimizer=SGD, epochs=1002021.0379.4
CE2021.0479.4
CE2022.0979.4
Cross Entropy2022.1079.4
CEBackbone=VGG-19bn, Number of Parameters=143.7M, FLOPs=19.7G2022.0779.4
ERM2022.0279.4
ERM2022.0279.4
Cross-Entropy2021.1179.4
CE2024.0879.4
BaselineBackbone=VGG19, Metadata availability=None2026.0379.4