paper-with-me

Data Augmentation 벤치마크

Data Augmentation on ImageNet

19개 결과 · ⬇ CSV · JSON

Accuracy (%)

73.8 76.08 78.35 80.62 82.9 2018-05 2026-09 ResNet-200 (AA) — 80.0 (2018-05-24) ResNet-50 (AA) — 77.6 (2018-05-24) ResNet-200 (Fast AA) — 80.6 (2019-05-01) ResNet-50 (Fast AA) — 77.6 (2019-05-01) ResNet-50 (RA) — 77.6 (2019-09-30) ResNet-50 (Faster AA) — 76.5 (2019-11-16) ResNet-50 (DADA) — 77.5 (2020-03-08) ResNet-200 (UA) — 80.4 (2020-03-31) ResNet-50 (UA) — 77.63 (2020-03-31) ResNet-50 (TA wide) — 78.07 (2021-03-18) ResNet-200 (DeepAA) — 81.32 (2022-03-11) ResNet-50 (DeepAA) — 78.3 (2022-03-11) ResNet-50 (LoRot-E) — 77.72 (2022-07-20) ResNet-50 (LoRot-I) — 77.71 (2022-07-20) DeiT-B (+MixPro) — 82.9 (2023-04-24) DeiT-S (+MixPro) — 81.3 (2023-04-24) DeiT-T (+MixPro) — 73.8 (2023-04-24) Sample-aware — 78.31 (2025-08-11) GeLDA — 74.7 (2026-02-02) ResNet-200 (AA) — 80.0 (2018-05-24) ResNet-200 (Fast AA) — 80.6 (2019-05-01) ResNet-200 (DeepAA) — 81.32 (2022-03-11) DeiT-B (+MixPro) — 82.9 (2023-04-24)
RankModel Accuracy (%) PaperCodeYear
1 DeiT-B (+MixPro) 82.9 MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer fistyee/mixpro 2023
2 ResNet-200 (DeepAA) 81.32 Deep AutoAugment msu-mlsys-lab/deepaa 2022
3 DeiT-S (+MixPro) 81.3 MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer fistyee/mixpro 2023
4 ResNet-200 (Fast AA) 80.6 Fast AutoAugment kakaobrain/fast-autoaugment · ildoonet/pytorch-randaugment · philip-bachman/amdim-public · +8 2019
5 ResNet-200 (UA) 80.4 UniformAugment: A Search-free Probabilistic Data Augmentation Approach tgilewicz/uniformaugment 2020
6 ResNet-200 (AA) 80.0 AutoAugment: Learning Augmentation Policies from Data tensorflow/models · tensorflow/models · pytorch/vision · +30 2018
7 Sample-aware 자동 추출 78.31 Sample-aware RandAugment: Search-free Automatic Data Augmentation for Effective Image Recognition 2025
8 ResNet-50 (DeepAA) 78.30 Deep AutoAugment msu-mlsys-lab/deepaa 2022
9 ResNet-50 (TA wide) 78.07 TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation pytorch/vision · automl/trivialaugment 2021
10 ResNet-50 (LoRot-E) 77.72 Tailoring Self-Supervision for Supervised Learning wjun0830/localizable-rotation 2022
11 ResNet-50 (LoRot-I) 77.71 Tailoring Self-Supervision for Supervised Learning wjun0830/localizable-rotation 2022
12 ResNet-50 (UA) 77.63 UniformAugment: A Search-free Probabilistic Data Augmentation Approach tgilewicz/uniformaugment 2020
13 ResNet-50 (AA) 77.6 AutoAugment: Learning Augmentation Policies from Data tensorflow/models · tensorflow/models · pytorch/vision · +30 2018
13 ResNet-50 (Fast AA) 77.6 Fast AutoAugment kakaobrain/fast-autoaugment · ildoonet/pytorch-randaugment · philip-bachman/amdim-public · +8 2019
13 ResNet-50 (RA) 77.6 RandAugment: Practical automated data augmentation with a reduced search space rwightman/pytorch-image-models · pytorch/vision · tensorflow/tpu · +16 2019
16 ResNet-50 (DADA) 77.5 DADA: Differentiable Automatic Data Augmentation VDIGPKU/DADA 2020
17 ResNet-50 (Faster AA) 76.5 Faster AutoAugment: Learning Augmentation Strategies using Backpropagation moskomule/dda 2019
18 GeLDA 자동 추출 74.7 Semantics-Aware Generative Latent Data Augmentation for Learning in Low-Resource Domains 2026
19 DeiT-T (+MixPro) 73.8 MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer fistyee/mixpro 2023
1–19 / 19 페이지당 10 20 50 100