paper-with-me

Action Classification 벤치마크

Action Classification on AViD

10개 결과 · ⬇ CSV · JSON

Accuracy

36.2 40.6 45 49.4 53.8 2020-07 2026-11 SlowFast-101 16x8 — 50.9 (2020-07-10) RepFlow ResNet-50 — 50.5 (2020-07-10) SlowFast-50 8x8 — 50.4 (2020-07-10) Two-Stream 3D ResNet-50 — 50.1 (2020-07-10) (2+1)D ResNet-50 — 48.8 (2020-07-10) SlowFast-50 4x4 — 48.5 (2020-07-10) 3D ResNet-50 — 48.2 (2020-07-10) I3D — 46.8 (2020-07-10) 2D ResNet-50 — 36.2 (2020-07-10) TokenLearner — 53.8 (2021-06-21) SlowFast-101 16x8 — 50.9 (2020-07-10) TokenLearner — 53.8 (2021-06-21)
RankModel Accuracy PaperCodeYear
1 TokenLearner 53.8 TokenLearner: What Can 8 Learned Tokens Do for Images and Videos? google-research/scenic · keras-team/keras-io · rish-16/tokenlearner-pytorch · +8 2021
2 SlowFast-101 16x8 50.9 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
3 RepFlow ResNet-50 50.5 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
4 SlowFast-50 8x8 50.4 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
5 Two-Stream 3D ResNet-50 50.1 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
6 (2+1)D ResNet-50 48.8 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
7 SlowFast-50 4x4 48.5 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
8 3D ResNet-50 48.2 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
9 I3D 46.8 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
10 2D ResNet-50 36.2 AViD Dataset: Anonymized Videos from Diverse Countries piergiaj/AViD 2020
1–10 / 10 페이지당 10 20 50 100