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/ResNet50

ResNet50

Reported on 34 benchmarks across 14 tasks · 18 papers · 20 SOTA

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

Computer Vision25 results

Self-Supervised Learningoncifar10
average top-1 classification accuracy· 2023-12-04
93.89
SOTA
Guarding Barlow Twins Against Overfitting with Mixed Samples arXiv:2312.02151
Self-Supervised Learningoncifar100
average top-1 classification accuracy· 2023-12-04
72.51
SOTA
Guarding Barlow Twins Against Overfitting with Mixed Samples arXiv:2312.02151
Self-Supervised LearningonTinyImageNet
average top-1 classification accuracy· 2023-12-04
51.84
SOTA
Guarding Barlow Twins Against Overfitting with Mixed Samples arXiv:2312.02151
Self-Supervised LearningonSTL-10
Accuracy· 2023-12-04
91.7
SOTA
Guarding Barlow Twins Against Overfitting with Mixed Samples arXiv:2312.02151
Image ClassificationonClothing1M (using clean data)
1:1 Accuracy· 2022-03-28
74.98
SOTA
UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning arXiv:2203.14542
Multi-Label Image ClassificationonBigEarthNet (official test set)
F1 Score· 2021-11-18
76.8
best: 80.8 (FG-MAE (ViT-S/16))
SOTA
Benchmarking and scaling of deep learning models for land cover image classification arXiv:2111.09451
Image ClassificationonBigEarthNet (official test set)
F1 Score· 2021-11-18
76.8
best: 80.8 (FG-MAE (ViT-S/16))
SOTA
Benchmarking and scaling of deep learning models for land cover image classification arXiv:2111.09451
Scene ClassificationonUC Merced Land Use Dataset
Accuracy (%)· 2019-11-15
99.61
best: 100 (µ2Net+ (ViT-L/16))
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Multi-Label Image ClassificationonBigEarthNet
mAP (macro)· 2019-11-15
75.36
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Image ClassificationonRESISC45
Top 1 Accuracy· uses extra data· 2019-11-15
96.83
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Image ClassificationonSo2Sat LCZ42
Accuracy· 2019-11-15
63.25
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Image ClassificationonEuroSAT
Accuracy (%)· uses extra data· 2019-11-15
99.2
best: 99.41 (DeepEnsembling)
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Image ClassificationonBigEarthNet
mAP (macro)· 2019-11-15
75.36
SOTA
In-domain representation learning for remote sensing arXiv:1911.06721
Parking Space OccupancyonSPKL
F1-score· 2023-06-07
0.6674
best: 0.7393 (EfficientNet-P)
Revising deep learning methods in parking lot occupancy detection arXiv:2306.04288
Parking Space OccupancyonACMPS
F1-score· 2023-06-07
0.9379
best: 0.9982 (EfficientNet-P)
Revising deep learning methods in parking lot occupancy detection arXiv:2306.04288
Parking Space OccupancyonAction-Camera Parking
F1-score· 2023-06-07
0.8377
best: 0.9343 (MobileNetV2)
Revising deep learning methods in parking lot occupancy detection arXiv:2306.04288
Parking Space OccupancyonPKLot
F1-score· 2023-06-07
0.9926
best: 0.9988 (VGG-19)
Revising deep learning methods in parking lot occupancy detection arXiv:2306.04288
Parking Space OccupancyonCNRPark+EXT
F1-score· 2023-06-07
0.938
best: 0.9683 (EfficientNet-P)
Revising deep learning methods in parking lot occupancy detection arXiv:2306.04288
Multi-Label Image ClassificationonBigEarthNet
FScore· 2021-11-18
76.8
best: 79 (WideResNet-B5-ECA)
Benchmarking and scaling of deep learning models for land cover image classification arXiv:2111.09451
Image ClassificationonBigEarthNet
FScore· 2021-11-18
76.8
best: 79 (WideResNet-B5-ECA)
Benchmarking and scaling of deep learning models for land cover image classification arXiv:2111.09451
Pose EstimationonCOCO test-dev
AP· 2020-08-17
73.7
best: 81.1 (ViTPose (ViTAE-G, ensemble))
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
Pose EstimationonCOCO minival
AP· 2020-08-17
75.3
best: 79.1 (HRNet-W48plus)
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
Multi-Person Pose EstimationonCOCO test-dev
AP· 2020-08-17
73.7
best: 79.2 (SCIO (HRNet-48))
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
Multi-Person Pose EstimationonCOCO minival
AP· 2020-08-17
75.3
best: 79.1 (HRNet-W48plus)
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
Image ClassificationonVTAB-1k
Top-1 Accuracy· 2019-10-01
42.1
best: 79.99 (ALIGN (50 hypers/task))
A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark arXiv:1910.04867

Methodology14 results

Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2023-05-27
73.6
SOTA
Matrix Information Theory for Self-Supervised Learning arXiv:2305.17326
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2020-05-20
73
best: 73.6
SOTA
What Makes for Good Views for Contrastive Learning?arXiv:2005.10243
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2020-03-09
71.1
best: 73.6
SOTA
Improved Baselines with Momentum Contrastive Learning arXiv:2003.04297
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2020-02-13
69.3
best: 73.6
SOTA
A Simple Framework for Contrastive Learning of Visual Representations arXiv:2002.05709
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2019-03-29
60.2
best: 73.6
SOTA
Local Aggregation for Unsupervised Learning of Visual Embeddings arXiv:1903.12355
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2018-05-05
56.5
best: 73.6
SOTA
Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination arXiv:1805.01978
Network PruningonImageNet
Accuracy· 2021-06-23
73.14
best: 78.79 (ResNet50-2.3 GFLOPs)
AC/DC: Alternating Compressed/DeCompressed Training of Deep Neural Networks arXiv:2106.12379
Network PruningonImageNet
Accuracy· 2021-05-07
75.59
best: 78.79 (ResNet50-2.3 GFLOPs)
Network Pruning That Matters: A Case Study on Retraining Variants arXiv:2105.03193
3DonCOCO test-dev
AP· 2020-08-17
73.7
best: 81.1 (ViTPose (ViTAE-G, ensemble))
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
3DonCOCO minival
AP· 2020-08-17
75.3
best: 79.1 (HRNet-W48plus)
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2019-12-04
63.6
best: 73.6
Self-Supervised Learning of Pretext-Invariant Representations arXiv:1912.01991
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2019-11-13
61.5
best: 73.6
Self-labelling via simultaneous clustering and representation learning arXiv:1911.05371
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2019-11-13
61.5
best: 73.6
Self-labelling via simultaneous clustering and representation learning arXiv:1911.05371
Contrastive Learningonimagenet-1k
ImageNet Top-1 Accuracy· 2019-11-13
60.6
best: 73.6
Momentum Contrast for Unsupervised Visual Representation Learning arXiv:1911.05722

Audio2 results

1 Image, 2*2 StitchionCOCO test-dev
AP· 2020-08-17
73.7
best: 81.1 (ViTPose (ViTAE-G, ensemble))
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139
1 Image, 2*2 StitchionCOCO minival
AP· 2020-08-17
75.3
best: 79.1 (HRNet-W48plus)
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation arXiv:2008.07139

Medical1 result

Document Text ClassificationonClothing1M (using clean data)
1:1 Accuracy· 2022-03-28
74.98
SOTA
UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning arXiv:2203.14542

Robots1 result

Activity RecognitiononUCF101
Accuracy 20%Test
98.05

Time Series1 result

Action RecognitiononUCF101
Accuracy 20%Test
98.05