Domain Adaptation on ImageNet-C

Metric: mean Corruption Error (mCE) (lower is better)

LeaderboardDataset

Loading chart...

Results

Hide augmentations

Sort:

#	Model↕	mean Corruption Error (mCE)▲	Augmentations	Paper	Date↕	Code
1	EfficientNet-L2+RPL	22	Yes	If your data distribution shifts, use self-learn...	2021-04-27	Code
2	EfficientNet-L2+ENT	23	Yes	If your data distribution shifts, use self-learn...	2021-04-27	Code
3	DINOv2 (ViT-g/14, frozen model, linear eval)	28.2	Yes	DINOv2: Learning Robust Visual Features without ...	2023-04-14	Code
4	CAFormer-B36 (IN21K, 384)	30.8	Yes	MetaFormer Baselines for Vision	2022-10-24	Code
5	MAE+DAT (ViT-H)	31.4	No	Enhance the Visual Representation via Discrete A...	2022-09-16	Code
6	DINOv2 (ViT-L/14, frozen model, linear eval)	31.5	Yes	DINOv2: Learning Robust Visual Features without ...	2023-04-14	Code
7	CAFormer-B36 (IN21K)	31.8	Yes	MetaFormer Baselines for Vision	2022-10-24	Code
8	MAE (ViT-H)	33.8	No	Masked Autoencoders Are Scalable Vision Learners	2021-11-11	Code
9	ResNeXt101 32x8d + DeepAug + Augmix + RPL	34.8	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
10	ConvFormer-B36 (IN21K)	35	Yes	MetaFormer Baselines for Vision	2022-10-24	Code
11	ResNeXt101 32x8d + DeepAug + Augmix + ENT	35.5	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
12	FAN-L-Hybrid (IN-22k)	35.8	Yes	Understanding The Robustness in Vision Transform...	2022-04-26	Code
13	Pyramid Adversarial Training Improves ViT (Im21k)	36.8	Yes	Pyramid Adversarial Training Improves ViT Perfor...	2021-11-30	Code
14	ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, full adaptation	38	No	Improving robustness against common corruptions ...	2020-06-30	Code
15	VOLO-D5+HAT	38.4	No	Improving Vision Transformers by Revisiting High...	2022-04-03	Code
16	DiscreteViT (Im21k)	38.74	Yes	Discrete Representations Strengthen Vision Trans...	2021-11-20	Code
17	ConvNeXt-XL (Im21k) (augmentation overlap with ImageNet-C)	38.8	Yes	A ConvNet for the 2020s	2022-01-10	Code
18	GPaCo (ViT-L)	39	No	Generalized Parametric Contrastive Learning	2022-09-26	Code
19	ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, 8 samples	40.7	No	Improving robustness against common corruptions ...	2020-06-30	Code
20	ResNeXt101 32x8d + IG-3.5B + ENT	40.8	Yes	If your data distribution shifts, use self-learn...	2021-04-27	Code
21	ResNeXt101 32x8d + IG-3.5B + RPL	40.9	Yes	If your data distribution shifts, use self-learn...	2021-04-27	Code
22	FAN-B-Hybrid (IN-22k)	41	Yes	Understanding The Robustness in Vision Transform...	2022-04-26	Code
23	Pyramid Adversarial Training Improves ViT	41.42	No	Pyramid Adversarial Training Improves ViT Perfor...	2021-11-30	Code
24	FAN-L-Hybrid+STL	42.1	No	Fully Attentional Networks with Self-emerging To...	2024-01-08	Code
25	QualNet (ResNeXt101)	42.5	No	-	-	Code
26	CAFormer-B36	42.6	No	MetaFormer Baselines for Vision	2022-10-24	Code
27	DINOv2 (ViT-B/14, frozen model, linear eval)	42.7	Yes	DINOv2: Learning Robust Visual Features without ...	2023-04-14	Code
28	FAN-L-Hybrid	43	No	Understanding The Robustness in Vision Transform...	2022-04-26	Code
29	ResNeXt101 32x8d + RPL	43.2	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
30	ResNeXt101 32x8d + ENT	44.3	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
31	ResNet50+DeepAug+AugMix, BatchNorm Adaptation, full adaptation	45.4	No	Improving robustness against common corruptions ...	2020-06-30	Code
32	DrViT	46.22	No	Discrete Representations Strengthen Vision Trans...	2021-11-20	Code
33	DiscreteViT	46.22	No	Discrete Representations Strengthen Vision Trans...	2021-11-20	Code
34	ConvFormer-B36	46.3	No	MetaFormer Baselines for Vision	2022-10-24	Code
35	RVT-B*	46.8	No	Towards Robust Vision Transformer	2021-05-17	Code
36	ResNet50+DeepAug+AugMix, BatchNorm Adaptation, 8 samples	48.4	No	Improving robustness against common corruptions ...	2020-06-30	Code
37	Sequencer2D-L	48.9	No	Sequencer: Deep LSTM for Image Classification	2022-05-04	Code
38	RVT-S*	49.4	No	Towards Robust Vision Transformer	2021-05-17	Code
39	ResNet-50 (PushPull-Conv) + PRIME	49.95	No	PushPull-Net: Inhibition-driven ResNet robust to...	2024-08-07	Code
40	ResNet50 + RPL	50.5	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
41	QualNet (ResNet-50)	50.6	No	-	-	Code
42	PRIME + DeepAugment (ResNet-50)	51.3	No	PRIME: A few primitives can boost robustness to ...	2021-12-27	Code
43	ResNet50 + ENT	51.6	No	If your data distribution shifts, use self-learn...	2021-04-27	Code
44	GFNet-S	53.8	No	Global Filter Networks for Image Classification	2021-07-01	Code
45	DINOv2 (ViT-S/14, frozen model, linear eval)	54.4	Yes	DINOv2: Learning Robust Visual Features without ...	2023-04-14	Code
46	PRIME with JSD (ResNet-50)	55.5	No	PRIME: A few primitives can boost robustness to ...	2021-12-27	Code
47	RVT-Ti*	57	No	Towards Robust Vision Transformer	2021-05-17	Code
48	PRIME (ResNet-50)	57.5	No	PRIME: A few primitives can boost robustness to ...	2021-12-27	Code
49	APR-SP + DeepAugment (ResNet-50)	57.5	No	Amplitude-Phase Recombination: Rethinking Robust...	2021-08-19	Code
50	DeepAugment (ResNet-50)	60.4	No	The Many Faces of Robustness: A Critical Analysi...	2020-06-29	Code
51	ResNet50 (baseline), BatchNorm Adaptation, full adaptation	62.2	No	Improving robustness against common corruptions ...	2020-06-30	Code
52	ResNet50 (baseline), BatchNorm Adaptation, 8 samples	65	No	Improving robustness against common corruptions ...	2020-06-30	Code
53	APR-SP (ResNet-50)	65	No	Amplitude-Phase Recombination: Rethinking Robust...	2021-08-19	Code
54	AugMix (ResNet-50)	65.3	No	AugMix: A Simple Data Processing Method to Impro...	2019-12-05	Code
55	Stylized ImageNet (ResNet-50)	69.3	Yes	ImageNet-trained CNNs are biased towards texture...	2018-11-29	Code
56	Group-wise Inhibition (ResNet-50)	69.6	No	Group-wise Inhibition based Feature Regularizati...	2021-03-03	Code
57	ResNet-50	76.7	No	Benchmarking Neural Network Robustness to Common...	2019-03-28	Code

#1EfficientNet-L2+RPLSOTA
22
mean Corruption Error (mCE)· Augmentations· 2021-04-27
If your data distribution shifts, use self-learning Code
#2EfficientNet-L2+ENTSOTA
23
mean Corruption Error (mCE)· Augmentations· 2021-04-27
If your data distribution shifts, use self-learning Code
#3DINOv2 (ViT-g/14, frozen model, linear eval)
28.2
mean Corruption Error (mCE)· Augmentations· 2023-04-14
DINOv2: Learning Robust Visual Features without Supervision Code
#4CAFormer-B36 (IN21K, 384)
30.8
mean Corruption Error (mCE)· Augmentations· 2022-10-24
MetaFormer Baselines for Vision Code
#5MAE+DAT (ViT-H)
31.4
mean Corruption Error (mCE)· 2022-09-16
Enhance the Visual Representation via Discrete Adversarial Training Code
#6DINOv2 (ViT-L/14, frozen model, linear eval)
31.5
mean Corruption Error (mCE)· Augmentations· 2023-04-14
DINOv2: Learning Robust Visual Features without Supervision Code
#7CAFormer-B36 (IN21K)
31.8
mean Corruption Error (mCE)· Augmentations· 2022-10-24
MetaFormer Baselines for Vision Code
#8MAE (ViT-H)
33.8
mean Corruption Error (mCE)· 2021-11-11
Masked Autoencoders Are Scalable Vision Learners Code
#9ResNeXt101 32x8d + DeepAug + Augmix + RPLSOTA
34.8
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#10ConvFormer-B36 (IN21K)
35
mean Corruption Error (mCE)· Augmentations· 2022-10-24
MetaFormer Baselines for Vision Code
#11ResNeXt101 32x8d + DeepAug + Augmix + ENTSOTA
35.5
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#12FAN-L-Hybrid (IN-22k)
35.8
mean Corruption Error (mCE)· Augmentations· 2022-04-26
Understanding The Robustness in Vision Transformers Code
#13Pyramid Adversarial Training Improves ViT (Im21k)
36.8
mean Corruption Error (mCE)· Augmentations· 2021-11-30
Pyramid Adversarial Training Improves ViT Performance Code
#14ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, full adaptationSOTA
38
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#15VOLO-D5+HAT
38.4
mean Corruption Error (mCE)· 2022-04-03
Improving Vision Transformers by Revisiting High-frequency Components Code
#16DiscreteViT (Im21k)
38.74
mean Corruption Error (mCE)· Augmentations· 2021-11-20
Discrete Representations Strengthen Vision Transformer Robustness Code
#17ConvNeXt-XL (Im21k) (augmentation overlap with ImageNet-C)
38.8
mean Corruption Error (mCE)· Augmentations· 2022-01-10
A ConvNet for the 2020s Code
#18GPaCo (ViT-L)
39
mean Corruption Error (mCE)· 2022-09-26
Generalized Parametric Contrastive Learning Code
#19ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, 8 samplesSOTA
40.7
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#20ResNeXt101 32x8d + IG-3.5B + ENT
40.8
mean Corruption Error (mCE)· Augmentations· 2021-04-27
If your data distribution shifts, use self-learning Code
#21ResNeXt101 32x8d + IG-3.5B + RPL
40.9
mean Corruption Error (mCE)· Augmentations· 2021-04-27
If your data distribution shifts, use self-learning Code
#22FAN-B-Hybrid (IN-22k)
41
mean Corruption Error (mCE)· Augmentations· 2022-04-26
Understanding The Robustness in Vision Transformers Code
#23Pyramid Adversarial Training Improves ViT
41.42
mean Corruption Error (mCE)· 2021-11-30
Pyramid Adversarial Training Improves ViT Performance Code
#24FAN-L-Hybrid+STL
42.1
mean Corruption Error (mCE)· 2024-01-08
Fully Attentional Networks with Self-emerging Token Labeling Code
#25QualNet (ResNeXt101)
42.5
mean Corruption Error (mCE)
No paperCode
#26CAFormer-B36
42.6
mean Corruption Error (mCE)· 2022-10-24
MetaFormer Baselines for Vision Code
#27DINOv2 (ViT-B/14, frozen model, linear eval)
42.7
mean Corruption Error (mCE)· Augmentations· 2023-04-14
DINOv2: Learning Robust Visual Features without Supervision Code
#28FAN-L-Hybrid
43
mean Corruption Error (mCE)· 2022-04-26
Understanding The Robustness in Vision Transformers Code
#29ResNeXt101 32x8d + RPL
43.2
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#30ResNeXt101 32x8d + ENT
44.3
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#31ResNet50+DeepAug+AugMix, BatchNorm Adaptation, full adaptationSOTA
45.4
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#32DrViT
46.22
mean Corruption Error (mCE)· 2021-11-20
Discrete Representations Strengthen Vision Transformer Robustness Code
#33DiscreteViT
46.22
mean Corruption Error (mCE)· 2021-11-20
Discrete Representations Strengthen Vision Transformer Robustness Code
#34ConvFormer-B36
46.3
mean Corruption Error (mCE)· 2022-10-24
MetaFormer Baselines for Vision Code
#35RVT-B*
46.8
mean Corruption Error (mCE)· 2021-05-17
Towards Robust Vision Transformer Code
#36ResNet50+DeepAug+AugMix, BatchNorm Adaptation, 8 samplesSOTA
48.4
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#37Sequencer2D-L
48.9
mean Corruption Error (mCE)· 2022-05-04
Sequencer: Deep LSTM for Image Classification Code
#38RVT-S*
49.4
mean Corruption Error (mCE)· 2021-05-17
Towards Robust Vision Transformer Code
#39ResNet-50 (PushPull-Conv) + PRIME
49.95
mean Corruption Error (mCE)· 2024-08-07
PushPull-Net: Inhibition-driven ResNet robust to image corruptions Code
#40ResNet50 + RPL
50.5
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#41QualNet (ResNet-50)
50.6
mean Corruption Error (mCE)
No paperCode
#42PRIME + DeepAugment (ResNet-50)
51.3
mean Corruption Error (mCE)· 2021-12-27
PRIME: A few primitives can boost robustness to common corruptions Code
#43ResNet50 + ENT
51.6
mean Corruption Error (mCE)· 2021-04-27
If your data distribution shifts, use self-learning Code
#44GFNet-S
53.8
mean Corruption Error (mCE)· 2021-07-01
Global Filter Networks for Image Classification Code
#45DINOv2 (ViT-S/14, frozen model, linear eval)
54.4
mean Corruption Error (mCE)· Augmentations· 2023-04-14
DINOv2: Learning Robust Visual Features without Supervision Code
#46PRIME with JSD (ResNet-50)
55.5
mean Corruption Error (mCE)· 2021-12-27
PRIME: A few primitives can boost robustness to common corruptions Code
#47RVT-Ti*
57
mean Corruption Error (mCE)· 2021-05-17
Towards Robust Vision Transformer Code
#48PRIME (ResNet-50)
57.5
mean Corruption Error (mCE)· 2021-12-27
PRIME: A few primitives can boost robustness to common corruptions Code
#49APR-SP + DeepAugment (ResNet-50)
57.5
mean Corruption Error (mCE)· 2021-08-19
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain Code
#50DeepAugment (ResNet-50)SOTA
60.4
mean Corruption Error (mCE)· 2020-06-29
The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization Code
#51ResNet50 (baseline), BatchNorm Adaptation, full adaptation
62.2
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#52ResNet50 (baseline), BatchNorm Adaptation, 8 samples
65
mean Corruption Error (mCE)· 2020-06-30
Improving robustness against common corruptions by covariate shift adaptation Code
#53APR-SP (ResNet-50)
65
mean Corruption Error (mCE)· 2021-08-19
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain Code
#54AugMix (ResNet-50)SOTA
65.3
mean Corruption Error (mCE)· 2019-12-05
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty Code
#55Stylized ImageNet (ResNet-50)SOTA
69.3
mean Corruption Error (mCE)· Augmentations· 2018-11-29
ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness Code
#56Group-wise Inhibition (ResNet-50)
69.6
mean Corruption Error (mCE)· 2021-03-03
Group-wise Inhibition based Feature Regularization for Robust Classification Code
#57ResNet-50
76.7
mean Corruption Error (mCE)· 2019-03-28
Benchmarking Neural Network Robustness to Common Corruptions and Perturbations Code