List your company
ResearchFormerly WizWand

Fine grained classification on Oxford-IIIT Pets

99.6Top-1 AccuracyOmniVec2
80.8885.7490.695.46Jan 17, 2020Feb 4, 2021Feb 24, 2022Mar 15, 2023Apr 3, 2024Apr 22, 2025May 12, 2026
Updated 6mo ago

Evaluation results

MethodLinks
OmniVec2Modality=Image2025.0799.6
OmniVecModality=Image2025.0799.2
DINOv3-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.0596.54
DINOv2-Reg-BEvaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.0595.97
AmoebaNet-BModel=AmoebaNet-B2020.0195.9
VECA-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 256, Core budget (C)=642026.0595.67
DINOv2-BEvaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.0595.61
VECA-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 256, Core budget (C)=82026.0595.12
Assemble-ResNet-FGVC-50Model=Assemble-ResNet-FGVC-502020.0194.3
ResNet-50 (ReLabel-trained)Backbone=ResNet-50, Pre-training strategy=ReLabel2021.0194.28
ResNet-50 (Baseline)Backbone=ResNet-50, Pre-training strategy=Standard2021.0193.92
DFN2B-CLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.0593.59
C-RADIOv3-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.0593.54
SigLIP2-B/16-256Evaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.0592.89
CLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.0592.72
ResNet-50Model=ResNet-502020.0192.5
OpenCLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.0591.91
CoCoOpPre-trained on ImageNet=true2024.1090.5
TransCLIPBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0889.3
CoOpPre-trained on ImageNet=true2024.1089.1
CLIP-VIT-B/16Pre-trained on ImageNet=false2024.1088.3
ADAPTBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0887.95
TPTPre-trained on ImageNet=false2024.1087.8
StatABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0887.7
DotaBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0887.33
EnsemblePre-trained on ImageNet=false2024.1086.9
DMNBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0886.78
ADAPTBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0886.43
TDABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0886.18
DPEBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0885.97
BCABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0885.58
TPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0884.49
TPSBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0884.4
C-TPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0884.1
CLIPBackbone=CLIP-RN50, BP-free=N/A, Evaluation Protocol=Online2025.0883.57
DiffTPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0883.4
APMPre-trained on ImageNet=false2024.1081.6