List your company
ResearchFormerly WizWand

Image Classification on Oxford-IIIT Pet

97.1AccuracyJFT - Adaptive Transfer
91.681693.088394.49595.9017Nov 16, 2018Feb 5, 2020Apr 27, 2021Jul 17, 2022Oct 7, 2023Dec 26, 2024Mar 18, 2026
Updated 7mo ago

Evaluation results

MethodLinks
JFT - Adaptive TransferPre-training Source=JFT, Backbone=Inception v3, Evaluation Protocol=Fine-tuning2018.1197.1——
JFT - AnimalPre-training Source=JFT, Subset=Animal, Backbone=Inception v3, Evaluation Protocol=Fine-tuning2018.1196.7——
DINOv2Arch.=ViT-L*, SSL Type Target=GL+DE, SSL Type Method=DIST+MIM, Evaluation Protocol=linear probing2026.0296.08——
CapPa L/14Backbone Scale=L/14, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0695.6——
DINOv2Arch.=ViT-B*, SSL Type Target=GL+DE, SSL Type Method=DIST+MIM, Evaluation Protocol=linear probing2026.0295.59——
CLIP L/14Backbone Scale=L/14, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0695.4——
SARELearning Paradigm=Training-based2026.0395.38——
CLIP* L/14Backbone Scale=L/14, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0695.2——
CrossViT-18Pre-training Dataset=ImageNet-1K, Evaluation Protocol=Fine-tuning2021.0395.07——
DeiT-SPre-training Dataset=ImageNet-1K, Evaluation Protocol=Fine-tuning2021.0394.93——
SFDAcomTop-k=k=32022.0794.92——
SFDATop-k=k=32022.0794.82——
NLEEPTop-k=k=32022.0794.71——
LogMETop-k=k=32022.0794.7——
AWTShots=162024.0794.63——
LogMETop-k=k=22022.0794.62——
SFDAcomTop-k=k=22022.0794.62——
NLEEPTop-k=k=22022.0794.59——
SFDATop-k=k=22022.0794.58——
CrossViT-15Pre-training Dataset=ImageNet-1K, Evaluation Protocol=Fine-tuning2021.0394.55——
CapPaBackbone Scale=B/16, Pre-training Batch Size=8k, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0694.4——
DeiT-BPre-training Dataset=ImageNet-1K, Evaluation Protocol=Fine-tuning2021.0394.39——
ImageNet - Adaptive TransferPre-training Source=ImageNet, Backbone=Inception v3, Evaluation Protocol=Fine-tuning2018.1194.1——
ProVP-RefShots=162024.0793.89——
DINOArch.=ViT-B, SSL Type Target=GLOBAL, SSL Type Method=DISTILL, Evaluation Protocol=linear probing2026.0293.84——
AWTShots=82024.0793.83——
PromptSRCShots=162024.0793.67——
CLIP* (16k)Backbone Scale=B/16, Pre-training Batch Size=16k, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0693.6——
CLIPBackbone Scale=B/16, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0693.6——
PLOT++Shots=162025.1093.6——
PLOT++Shots=162024.0793.59——
ISyNet-N1-S32021.0993.57——
ISyNet-N22021.0993.55——
PromptSRCShots=82024.0793.5——
AWTShots=42024.0793.5——
ISyNet-N32021.0993.46——
APE-TShots=82024.0793.46——
CoCoOpShots=82024.0793.45——
PromptSRCShots=42024.0793.43——
CLIP* (8k)Backbone Scale=B/16, Pre-training Batch Size=8k, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0693.4——
CoCoOpShots=162024.0793.34——
ImageNet - Entire DatasetPre-training Source=ImageNet, Backbone=Inception v3, Evaluation Protocol=Fine-tuning2018.1193.3——
ResNet-34+2021.0993.3——
ProVP-RefShots=82024.0793.27——
ResNet-50+2021.0993.22——
SupervisedProtocol=Fine-tuned, Backbone=ResNet-50 (4x), Pre-training=ImageNet2020.0293.2——
CoCoOpShots=162025.1093.2——
KgCoOpShots=162025.1093.2——
APE-TShots=162024.0793.16——
CapBackbone Scale=B/16, Pre-training Batch Size=8k, Evaluation Protocol=Frozen visual representations with single transformer decoder2023.0693.1——
LRWOpt2024.0393.09——
PLOT++Shots=82024.0793.02——
ProVP-RefShots=42024.0793.02——
Supervised(IN1K)Pretrain Dataset=1.2M, Backbone=ResNet-50, Evaluation Protocol=End-to-end fine-tuning2022.0493——
AWTShots=22024.0792.98——
AWTShots=12024.0792.89——
CoCoOpShots=42024.0792.81——
ProGradShots=162025.1092.8——
LRW-Hard2024.0392.72——
CoCoOpShots=42025.1092.7——
PLOT++Shots=42025.1092.7——
ISyNet-N1-S22021.0992.65——
CoCoOpShots=22024.0792.64——
KgCoOpShots=42025.1092.6——
TIP-AdapterShots=162025.1092.6——
CoOpShots=42024.0792.57——
PLOT++Shots=42024.0792.55——
STELLARArch.=ViT-L, SSL Type Target=SPARSE, SSL Type Method=INV+REC, Evaluation Protocol=linear probing2026.0292.53——
STELLARArch.=ViT-H, SSL Type Target=SPARSE, SSL Type Method=INV+REC, Evaluation Protocol=linear probing2026.0292.53——
FSR2024.0392.51——
Entire JFT DatasetPre-training Source=JFT, Backbone=Inception v3, Evaluation Protocol=Fine-tuning2018.1192.5——
PromptSRCShots=22024.0792.5——
CoOpShots=42025.1092.5——
iBOTArch.=ViT-B, SSL Type Target=GL+DE, SSL Type Method=DIST+MIM, Evaluation Protocol=linear probing2026.0292.4——
LRW-Random2024.0392.38——
BILAW2024.0392.38——
ISyNet-N1-S12021.0992.35——
RHO-LOSS2024.0392.35——
APE-TShots=42024.0792.34——
ERM2024.0392.33——
ProVP-RefShots=22024.0792.31——
SupervisedProtocol=Linear evaluation, Backbone=ResNet-50 (4x), Pre-training=ImageNet2020.0292.3——
CLIP-AdapterShots=162025.1092.3——
PLOT++Shots=22024.0792.29——
IndividualBackbone=ViT-B/16, Pre-trained=ImageNet-21k, Number of Merged Models=12024.0592.23——
MBW2024.0392.18——
MAPLE2024.0392.17——
iBOTArch.=ViT-L, SSL Type Target=GL+DE, SSL Type Method=DIST+MIM, Evaluation Protocol=linear probing2026.0292.12——
SupervisedEvaluation protocol=Fine-tuned, Pre-training=ImageNet, Backbone=ResNet-502020.0292.1——
SimCLRProtocol=Fine-tuned, Backbone=ResNet-50 (4x), Pre-training=ImageNet2020.0292.1——
KgCoOpShots=12025.1092.1——
CLIP-LoRA + FMAShots=12025.1092.1——
ProGradShots=42025.1092.1——
MWN2024.0392.04——
PromptSRCShots=12024.0792——
CoCoOpShots=12025.1091.9——
PLOT++Shots=12025.1091.9——
CLIP-LoRAShots=12025.1091.9——
TIP-AdapterShots=42025.1091.9——
PLOT++Shots=12024.0791.89——