paper-with-me

Few-Shot Image Classification 벤치마크

Few-Shot Image Classification on CUB 200 5-way 1-shot

72개 결과 · ⬇ CSV · JSON

Accuracy

50.44 61.78 73.12 84.46 95.8 2017-11 2026-09 Relation Net — 50.44 (2017-11-16) Relation Net — 50.44 (2017-11-16) Delta-encoder — 69.8 (2018-06-12) Delta-encoder — 69.8 (2018-06-12) feat (ProtoNet) — 68.65 (2018-12-10) feat (ProtoNet) — 68.65 (2018-12-10) DN4-DA (k=1) — 53.15 (2019-03-28) DN4-DA (k=1) — 53.15 (2019-03-28) Hyperbolic ProtoNet — 60.52 (2019-04-03) Hyperbolic ProtoNet — 60.52 (2019-04-03) Self-Critique and Adapt + High-End MAML++ — 70.46 (2019-05-24) High-End MAML++ — 67.48 (2019-05-24) Self-Critique and Adapt + High-End MAML++ — 70.46 (2019-05-24) High-End MAML++ — 67.48 (2019-05-24) S2M2R — 80.68 (2019-07-28) S2M2R — 80.68 (2019-07-28) DKT + BNCosSim — 72.27 (2019-10-11) DKT + BNCosSim — 72.27 (2019-10-11) AmdimNet — 77.09 (2019-11-14) AmdimNet — 77.09 (2019-11-14) Transfer+SGC — 88.35 (2020-01-27) Transfer+SGC — 88.35 (2020-01-27) ICI — 89.58 (2020-03-26) Neg-Margin — 72.66 (2020-03-26) ICI — 89.58 (2020-03-26) Neg-Margin — 72.66 (2020-03-26) PT+MAP — 91.55 (2020-06-06) PT+MAP — 91.55 (2020-06-06) LaplacianShot — 80.96 (2020-06-29) LaplacianShot — 80.96 (2020-06-29) TIM-GD — 82.2 (2020-08-25) TIM-GD — 82.2 (2020-08-25) Ours — 79.12 (2020-10-07) Ours — 79.12 (2020-10-07) Illumination Augmentation — 94.73 (2021-02-06) Illumination Augmentation — 94.73 (2021-02-06) LST+MAP — 91.68 (2021-02-09) LST+MAP — 91.68 (2021-02-09) RS-FSL — 65.66 (2021-04-26) RS-FSL — 65.66 (2021-04-26) PT+MAP+SF (transductive) — 95.48 (2021-05-11) PT+MAP+SF (transductive) — 95.48 (2021-05-11) BD-CSPN + ESFR (ResNet-18) — 82.68 (2021-06-22) BD-CSPN + ESFR (ResNet-18) — 82.68 (2021-06-22) RENet — 79.49 (2021-08-22) RENet — 79.49 (2021-08-22) PEMnE-BMS* — 94.78 (2021-10-18) PEMnE-BMS* — 94.78 (2021-10-18) EASY 3xResNet12 (transductive) — 90.56 (2022-01-24) EASY 4xResNet12 (transductive) — 90.5 (2022-01-24) EASY 3xResNet12 (inductive) — 78.56 (2022-01-24) EASY 4xResNet12 (inductive) — 77.97 (2022-01-24) EASY 3xResNet12 (transductive) — 90.56 (2022-01-24) EASY 4xResNet12 (transductive) — 90.5 (2022-01-24) EASY 3xResNet12 (inductive) — 78.56 (2022-01-24) EASY 4xResNet12 (inductive) — 77.97 (2022-01-24) HyperShot — 66.13 (2022-03-21) HyperShot — 66.13 (2022-03-21) PT+MAP+SF+SOT (transductive) — 95.8 (2022-04-06) PT+MAP+SF+SOT (transductive) — 95.8 (2022-04-06) TDM — 84.36 (2022-07-04) TDM — 84.36 (2022-07-04) BAVARDAGE — 90.42 (2022-09-18) BAVARDAGE — 90.42 (2022-09-18) MergedNet-Max — 75.34 (2022-10-14) MergedNet-Max — 75.34 (2022-10-14) ESPT — 85.45 (2023-04-26) ESPT — 85.45 (2023-04-26) CAML [Laion-2b] — 95.8 (2023-10-17) CAML [Laion-2b] — 95.8 (2023-10-17) PT+MAP+SF+BPA (transductive) — 95.8 (2024-06-25) PT+MAP+SF+BPA (transductive) — 95.8 (2024-06-25) Relation Net — 50.44 (2017-11-16) Delta-encoder — 69.8 (2018-06-12) Self-Critique and Adapt + High-End MAML++ — 70.46 (2019-05-24) S2M2R — 80.68 (2019-07-28) Transfer+SGC — 88.35 (2020-01-27) ICI — 89.58 (2020-03-26) PT+MAP — 91.55 (2020-06-06) Illumination Augmentation — 94.73 (2021-02-06) PT+MAP+SF (transductive) — 95.48 (2021-05-11) PT+MAP+SF+SOT (transductive) — 95.8 (2022-04-06)
RankModel Accuracy Extra Training Data PaperCodeYear
1 PT+MAP+SF+SOT (transductive) 95.80 The Self-Optimal-Transport Feature Transform danielshalam/bpa 2022
1 CAML [Laion-2b] 95.8 Context-Aware Meta-Learning cfifty/CAML 2023
1 PT+MAP+SF+BPA (transductive) 95.80 The Balanced-Pairwise-Affinities Feature Transform danielshalam/bpa 2024
4 PT+MAP+SF (transductive) 95.48 Few-Shot Learning by Integrating Spatial and Frequency Representation xiangyu8/PT-MAP-sf 2021
5 PEMnE-BMS* 94.78 Squeezing Backbone Feature Distributions to the Max for Efficient Few-Shot Learning yhu01/bms 2021
6 Illumination Augmentation 94.73 Sill-Net: Feature Augmentation with Separated Illumination Representation lanfenghuanyu/Sill-Net 2021
7 LST+MAP 91.68 Transfer learning based few-shot classification using optimal transport mapping from preprocessed latent space of backbone neural network ctom2/latent-space-transform 2021
8 PT+MAP 91.55% Leveraging the Feature Distribution in Transfer-based Few-Shot Learning sicara/easy-few-shot-learning · yhu01/PT-MAP · yhu01/bms · +3 2020
9 EASY 3xResNet12 (transductive) 90.56 EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients ybendou/easy · brain-bzh/pefsl · ybendou/fs-generalization 2022
10 EASY 4xResNet12 (transductive) 90.5 EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients ybendou/easy · brain-bzh/pefsl · ybendou/fs-generalization 2022
1–10 / 72 다음 → 페이지당 10 20 50 100