List your company
ResearchFormerly WizWand

Fine-grained visual classification on CUB-200-2011 (test)

0.918Top-1 AccMP-FGVC
0.401640.5356950.669750.803805Sep 16, 2023Feb 9, 2024Jul 4, 2024Nov 27, 2024Apr 22, 2025Sep 15, 2025Feb 8, 2026
Updated 8mo ago

Evaluation results

MethodLinks
MP-FGVCBackbone=ViT-B-162023.090.918
SIM-TransBackbone=ViT-B-16, reproduced_from_source=true2023.090.913
TransFGBackbone=ViT-B-16, reproduced_from_source=true2023.090.911
IELTBackbone=ViT-B-16, reproduced_from_source=true2023.090.909
CALBackbone=CNN2023.090.906
API-NetBackbone=CNN2023.090.9
PRISBackbone=CNN2023.090.9
RLRR*Backbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.47, AugReg=true2024.030.898
SSFBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.39, AugReg=true2024.030.895
RLRRBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.47, AugReg=false2024.030.893
ARC*Backbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.25, AugReg=true2024.030.893
FDLBackbone=CNN2023.090.891
MSHQPBackbone=CNN2023.090.89
MRDMNBackbone=CNN2023.090.888
VPT-DeepBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.85, AugReg=false2024.030.885
ARCBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.25, AugReg=false2024.030.885
BiasBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.28, AugReg=false2024.030.884
LoRABackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.44, AugReg=false2024.030.883
Full fine-tuningBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=85.98, AugReg=false2024.030.873
AdapterBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.41, AugReg=false2024.030.871
VPT-ShallowBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.25, AugReg=false2024.030.867
Linear probingBackbone=ViT-B/16, Pre-training=ImageNet-21k, Params. (M)=0.18, AugReg=false2024.030.853
SPT-DeepBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K, Tuning Depth=Deep2024.020.8447
SPT-ShallowBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K, Tuning Depth=Shallow2024.020.835
SaSPAType=Ours, Backbone=ResNet-50, Diffusion Model=BLIP-diffusion2024.060.832
SaSPA w/o BLIP-diffusionType=Ours, Backbone=ResNet-50, Diffusion Model=Stable Diffusion v1.52024.060.83
GateVPTBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K2024.020.8286
Real GuidanceType=Generative, Backbone=ResNet-502024.060.828
VPT-DeepBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K, Tuning Depth=Deep2024.020.8267
CAL-AugType=Traditional, Backbone=ResNet-502024.060.825
CAL-Aug + CutMixType=Traditional, Backbone=ResNet-502024.060.824
ALIAType=Generative, Backbone=ResNet-502024.060.82
CutMixType=Traditional, Backbone=ResNet-502024.060.818
FullBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K, Tuning Depth=Full2024.020.8175
No AugType=Traditional, Backbone=ResNet-502024.060.815
RandAugType=Traditional, Backbone=ResNet-502024.060.815
RandAug + CutMixType=Traditional, Backbone=ResNet-502024.060.812
QuICBackbone=VGG16, Resolution=448x4482026.010.81
FullBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K, Tuning Depth=Full2024.020.8055
QuICBackbone=ResNet18, Resolution=448x4482026.010.805
QuICBackbone=GoogLeNet, Resolution=448x4482026.010.803
SPT-DeepBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K, Tuning Depth=Deep2024.020.8013
VPT-ShallowBackbone=ViT-B, Pre-training Method=MoCo-v3, Pre-training Dataset=ImageNet-1K, Tuning Depth=Shallow2024.020.7905
GAPBackbone=ResNet18, Resolution=448x4482026.010.788
GAPBackbone=GoogLeNet, Resolution=448x4482026.010.784
SE-BlockBackbone=ResNet18, Resolution=448x4482026.010.777
GAPBackbone=VGG16, Resolution=448x4482026.010.77
FCBackbone=GoogLeNet, Resolution=448x4482026.010.767
SE-BlockBackbone=VGG16, Resolution=448x4482026.010.766
PANDStudent=MobileNet-V22026.020.7652
PANDStudent=ResNet-182026.020.7609
SE-BlockBackbone=GoogLeNet, Resolution=448x4482026.010.756
VL2LiteStudent=ResNet-182026.020.7267
VL2LiteStudent=MobileNet-V22026.020.7219
SPT-ShallowBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K, Tuning Depth=Shallow2024.020.7115
KDStudent=ResNet-182026.020.7095
GateVPTBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K2024.020.7056
VPT-DeepBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K, Tuning Depth=Deep2024.020.6833
RKDStudent=ResNet-182026.020.6831
KDStudent=MobileNet-V22026.020.68
RKDStudent=MobileNet-V22026.020.6795
w/o KDStudent=MobileNet-V22026.020.6542
w/o KDStudent=ResNet-182026.020.6448
FCBackbone=ResNet18, Resolution=448x4482026.010.635
QuICBackbone=AlexNet, Resolution=448x4482026.010.622
FCBackbone=VGG16, Resolution=448x4482026.010.613
GAPBackbone=AlexNet, Resolution=448x4482026.010.584
SE-BlockBackbone=AlexNet, Resolution=448x4482026.010.579
FCBackbone=AlexNet, Resolution=448x4482026.010.524
VPT-ShallowBackbone=ViT-B, Pre-training Method=MAE, Pre-training Dataset=ImageNet-1K, Tuning Depth=Shallow2024.020.4215