List your company
ResearchFormerly WizWand

Image Classification on Stanford Cars (val)

94.5AccuracyDeiT III-L
-1.38823.50648.473.294Oct 15, 2019Nov 6, 2020Nov 30, 2021Dec 23, 2022Jan 16, 2024Feb 7, 2025Mar 3, 2026
Updated 8mo ago

Evaluation results

MethodLinks
DeiT III-LBackbone=ViT-L, Pre-training=ImageNet-1k, Epochs=400, Resolution=224x224, Crop ratio=1.02022.0494.5
DeiT III-BBackbone=ViT-B, Pre-training=ImageNet-1k, Epochs=400, Resolution=224x224, Crop ratio=1.02022.0493.4
CPGBackbone=ResNet-50, Model Size (MB)=1212019.1092.8
SigLIxPTrain-free=true, Shots=16, Backbone=ViT-B/162024.1192.8
Grafit ResNet-50 [52]Pre-training=ImageNet-1k, Crop ratio=0.8752022.0492.5
DeiT-BPre-training=ImageNet-1k, Crop ratio=0.8752022.0492.1
FinetuneBackbone=ResNet-50, Model Size (MB)=5542019.1091.83
DeiT III-SBackbone=ViT-S, Pre-training=ImageNet-1k, Epochs=400, Resolution=224x224, Crop ratio=1.02022.0489.9
PiggybackBackbone=ResNet-50, Model Size (MB)=1212019.1089.62
ProgressiveNetBackbone=ResNet-50, Model Size (MB)=5632019.1089.21
MentorMixtraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1186.9
CasPLTrain-free=false, Shots=16, Backbone=ViT-B/162024.1186.7
PackNetBackbone=ResNet-50, Model Size (MB)=1152019.1086.11
DMNTrain-free=false, Shots=16, Backbone=ViT-B/162024.1185.3
Mixuptraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1185
WeightDecaytraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1184.3
Dropouttraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1184.1
Dropouttraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1183.8
PromptSRCTrain-free=false, Shots=16, Backbone=ViT-B/162024.1183.8
MentorMixtraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1183.6
MaPLeTrain-free=false, Shots=16, Backbone=ViT-B/162024.1183.6
Bootstraptraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1182.8
MentorNettraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1182.6
Mixuptraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1182.5
Vanillatraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1182.4
S-Modeltraining_setting=Fine-tuned, noise_type=Red, backbone=Inception-ResNet-v22019.1182.4
Linear ProbeTrain-free=false, Shots=16, Backbone=ViT-B/162024.1180.4
Bootstraptraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1178.9
MentorNettraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1178.9
DMN-TFTrain-free=true, Shots=16, Backbone=ViT-B/162024.1178.4
MentorMixtraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1178.2
Vanillatraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1177.7
S-Modeltraining_setting=Trained from scratch, noise_type=Red, backbone=Inception-ResNet-v22019.1177.6
MentorNettraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1175.9
Mixuptraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1173.1
WeightDecaytraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1172.2
Dropouttraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1171.7
Bootstraptraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1171.7
Vanillatraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1170
S-Modeltraining_setting=Fine-tuned, noise_type=Blue, backbone=Inception-ResNet-v22019.1169.7
MentorMixtraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1167.7
OTTEREvaluation Protocol=Linear Probing, Label Distribution Estimation=BBSE+OT2024.0466.3
BBSE+PMEvaluation Protocol=Linear Probing, Label Distribution Estimation=BBSE+PM2024.0465.4
Linear Probing baselineEvaluation Protocol=Linear Probing, Label Distribution Estimation=None2024.0464.5
Mixuptraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1164.2
Dropouttraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1162.8
Train from ScratchBackbone=ResNet-50, Model Size (MB)=5542019.1061.56
Reference Masking RegularizationBackbone=ViTb16, Pre-trained=ImageNet-1K, Adaptation Protocol=Linear Probing2024.0661.15
StandardBackbone=ViTb16, Pre-trained=ImageNet-1K, Adaptation Protocol=Linear Probing2024.0658.54
MentorNettraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1156.8
Zero-Shot baselineEvaluation Protocol=Zero-Shot, Label Distribution Estimation=None2024.0455.7
Bootstraptraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1155.6
S-Modeltraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1153.9
Vanillatraining_setting=Trained from scratch, noise_type=Blue, backbone=Inception-ResNet-v22019.1153.8
OTTEREvaluation Protocol=Zero-Shot, Label Distribution Estimation=BBSE+OT2024.0451.7
ITO_sub3Pre-training Dataset=CC3M-recap, Backbone=ViT-B/16, Evaluation Protocol=Linear Probing2026.0325.32
FLAIRPre-training Dataset=CC3M-recap, Backbone=ViT-B/16, Evaluation Protocol=Linear Probing2026.0325.05
ITO_sub2Pre-training Dataset=CC3M-recap, Backbone=ViT-B/16, Evaluation Protocol=Linear Probing2026.0323.96
BBSE+PMEvaluation Protocol=Zero-Shot, Label Distribution Estimation=BBSE+PM2024.042.3