Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/ResNet-50

ResNet-50

Reported on 55 benchmarks across 19 tasks · 10 papers · 24 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision22 results

Image ClassificationoniNaturalist 2018
Top-1 Accuracy· 2020-06-17
48.6
best: 94.6 (OmniVec2)
SOTA
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments arXiv:2006.09882
Pose EstimationonCOCO (Common Objects in Context)
Validation AP· 2018-04-17
72.2
best: 82.2 (Sapiens-2B)
SOTA
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Period EstimationonOmniArt
Mean absolute error· 2017-08-02
79.3
best: 77.9 (OmniArt)
SOTA
OmniArt: Multi-task Deep Learning for Artistic Data Analysis arXiv:1708.00684
Image ClassificationonGasHisSDB
Accuracy· 2015-12-10
98.56
best: 98.74 (CoAtNet-1)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Image ClassificationonGasHisSDB
F1-Score· 2015-12-10
99.24
best: 99.38 (CoAtNet-1)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Image ClassificationonGasHisSDB
Precision· 2015-12-10
99.94
best: 99.97 (CoAtNet-1)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain GeneralizationonImageNet-R
Top-1 Error Rate· 2015-12-10
63.9
best: 3.9 (Model soups (BASIC-L))
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Image ClassificationonCUB-200-2011
Accuracy· 2023-08-25
88.59
best: 92.8 (PIM)
PCNN: Probable-Class Nearest-Neighbor Explanations Improve Fine-Grained Image Classification Accuracy for AIs and Humans arXiv:2308.13651
Fine-Grained Image ClassificationonCUB-200-2011
Accuracy· 2023-08-25
88.59
best: 92.8 (PIM)
PCNN: Probable-Class Nearest-Neighbor Explanations Improve Fine-Grained Image Classification Accuracy for AIs and Humans arXiv:2308.13651
Domain GeneralizationonImageNet-C
mean Corruption Error (mCE)· 2019-03-28
76.7
best: 28.2 (DINOv2 (ViT-g/14, frozen model, linear eval))
Benchmarking Neural Network Robustness to Common Corruptions and Perturbations arXiv:1903.12261
Object Recognitiononshape bias
shape bias· 2018-11-29
22.1
best: 98.7 (Imagen)
ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness arXiv:1811.12231
Pose EstimationonOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Pose EstimationonOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Pose EstimationonOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Pose EstimationonOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Image ClassificationonOmniBenchmark
Average Top-1 Accuracy· 2015-12-10
34.3
best: 47.6 (NOAH-ViTB/16)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Image ClassificationonImageNet
GFLOPs· 2015-12-10
3.8
best: 1478 (InternImage-H)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain GeneralizationonVizWiz-Classification
Accuracy - All Images· 2015-12-10
42.9
best: 57.2 (VOLO-D5)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain GeneralizationonVizWiz-Classification
Accuracy - Clean Images· 2015-12-10
47.7
best: 450 (ViT-8/B-224)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain GeneralizationonVizWiz-Classification
Accuracy - Corrupted Images· 2015-12-10
37.1
best: 51.8 (VOLO-D5)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Image ClassificationonTCMP-300
Accuracy (% )
87.47
best: 89.64 (Swin-Base)
Domain GeneralizationonImageNet-A
Top-1 accuracy %
0
best: 94.17 (Model soups (BASIC-L))

Methodology21 results

ClassificationonICBHI Respiratory Sound Database
ICBHI Score· uses extra data· 2021-08-04
58.29
best: 65.53 (ADD)
SOTA
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
ClassificationonICBHI Respiratory Sound Database
Specificity· uses extra data· 2021-08-04
79.34
best: 85.13 (ADD)
SOTA
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
3DonCOCO (Common Objects in Context)
Validation AP· 2018-04-17
72.2
best: 82.2 (Sapiens-2B)
SOTA
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Domain AdaptationonOffice-31
Average Accuracy· 2015-12-10
76.1
best: 96 (FFTAT)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain AdaptationonImageNet-R
Top-1 Error Rate· 2015-12-10
63.9
best: 3.9 (Model soups (BASIC-L))
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
ClassificationonNCT-CRC-HE-100K
Accuracy (%)· 2015-12-10
94.72
best: 95.59 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
ClassificationonNCT-CRC-HE-100K
F1-Score· 2015-12-10
97.09
best: 97.48 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
ClassificationonNCT-CRC-HE-100K
Precision· 2015-12-10
100
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
ClassificationonNCT-CRC-HE-100K
Specificity· 2015-12-10
99.34
best: 99.45 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Remote SensingonFireRisk
Accuracy (%)· 2023-03-13
63.2
best: 65.29 (MAE (ViT-B/16))
FireRisk: A Remote Sensing Dataset for Fire Risk Assessment with Benchmarks Using Supervised and Self-supervised Learning arXiv:2303.07035
ClassificationonICBHI Respiratory Sound Database
Sensitivity· uses extra data· 2021-08-04
37.24
best: 48.21 (BEATs (CE))
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
Zero-Shot LearningonCOCO-MLT
Average mAP· uses extra data· 2021-02-26
56.19
best: 60.17 (ViT-B/16)
Learning Transferable Visual Models From Natural Language Supervision arXiv:2103.00020
Domain AdaptationonImageNet-C
mean Corruption Error (mCE)· 2019-03-28
76.7
best: 22 (EfficientNet-L2+RPL)
Benchmarking Neural Network Robustness to Common Corruptions and Perturbations arXiv:1903.12261
3DonOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
3DonOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
3DonOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
3DonOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Domain AdaptationonVizWiz-Classification
Accuracy - All Images· 2015-12-10
42.9
best: 57.2 (VOLO-D5)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain AdaptationonVizWiz-Classification
Accuracy - Clean Images· 2015-12-10
47.7
best: 450 (ViT-8/B-224)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain AdaptationonVizWiz-Classification
Accuracy - Corrupted Images· 2015-12-10
37.1
best: 51.8 (VOLO-D5)
Deep Residual Learning for Image Recognition arXiv:1512.03385
Domain AdaptationonImageNet-A
Top-1 accuracy %
0
best: 94.17 (Model soups (BASIC-L))

Audio9 results

Audio ClassificationonICBHI Respiratory Sound Database
ICBHI Score· uses extra data· 2021-08-04
58.29
best: 65.53 (ADD)
SOTA
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
Audio ClassificationonICBHI Respiratory Sound Database
Specificity· uses extra data· 2021-08-04
79.34
best: 85.13 (ADD)
SOTA
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
1 Image, 2*2 StitchionCOCO (Common Objects in Context)
Validation AP· 2018-04-17
72.2
best: 82.2 (Sapiens-2B)
SOTA
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
Audio ClassificationonICBHI Respiratory Sound Database
Sensitivity· uses extra data· 2021-08-04
37.24
best: 48.21 (BEATs (CE))
Lung Sound Classification Using Co-tuning and Stochastic Normalization arXiv:2108.01991
1 Image, 2*2 StitchionOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
1 Image, 2*2 StitchionOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
1 Image, 2*2 StitchionOCHuman
Test AP· 2018-04-17
29.5
best: 93.3 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
1 Image, 2*2 StitchionOCHuman
Validation AP· 2018-04-17
32.1
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
10-shot image generationonDADA-seg
mIoU· 2015-12-10
18.96
best: 46.97 (MMUDA)
Deep Residual Learning for Image Recognition arXiv:1512.03385

Medical5 results

Medical Image ClassificationonNCT-CRC-HE-100K
Accuracy (%)· 2015-12-10
94.72
best: 95.59 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Medical Image ClassificationonNCT-CRC-HE-100K
F1-Score· 2015-12-10
97.09
best: 97.48 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Medical Image ClassificationonNCT-CRC-HE-100K
Precision· 2015-12-10
100
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Medical Image ClassificationonNCT-CRC-HE-100K
Specificity· 2015-12-10
99.34
best: 99.45 (Efficientnet-b0)
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385
Semantic SegmentationonDADA-seg
mIoU· 2015-12-10
18.96
best: 46.97 (MMUDA)
Deep Residual Learning for Image Recognition arXiv:1512.03385

Knowledge Base2 results

2D Human Pose EstimationonOCHuman
Test AP· 2018-04-17
30.4
best: 48.3 (BBox-Mask-Pose 2x)
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208
2D Human Pose EstimationonOCHuman
Validation AP· 2018-04-17
37.8
best: 48.6 (BBox-Mask-Pose 2x)
Simple Baselines for Human Pose Estimation and Tracking arXiv:1804.06208

Speech1 result

Speaker VerificationonVoxCeleb2
EER· uses extra data· 2015-12-10
100
SOTA
Deep Residual Learning for Image Recognition arXiv:1512.03385

Computer Code1 result

Remote Sensing Image ClassificationonFireRisk
Accuracy (%)· 2023-03-13
63.2
best: 65.29 (MAE (ViT-B/16))
FireRisk: A Remote Sensing Dataset for Fire Risk Assessment with Benchmarks Using Supervised and Self-supervised Learning arXiv:2303.07035