Action Anticipation On Epic Kitchens 55 Seen
Métriques
Top 1 Accuracy - Act.
Top 1 Accuracy - Noun
Top 1 Accuracy - Verb
Top 5 Accuracy - Act.
Top 5 Accuracy - Noun
Top 5 Accuracy - Verb
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Top 1 Accuracy - Act. | Top 1 Accuracy - Noun | Top 1 Accuracy - Verb | Top 5 Accuracy - Act. | Top 5 Accuracy - Noun | Top 5 Accuracy - Verb | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|
| AVT+ | 16.84 | 20.16 | 34.36 | 36.52 | 51.57 | 80.03 | Anticipative Video Transformer | |
| Abstract Goal | 22.03 | 51.56 | 35.34 | 38.29 | 58.01 | 82.56 | Predicting the Next Action by Modeling the Abstract Goal | - |
| 2SCNN | 4.32 | 15.15 | 29.76 | 15.21 | 38.56 | 76.03 | Scaling Egocentric Vision: The EPIC-KITCHENS Dataset | |
| ATSN | 6.00 | 16.22 | 31.81 | 28.21 | 42.15 | 76.56 | Scaling Egocentric Vision: The EPIC-KITCHENS Dataset | |
| ImagineRNN | 14.66 | 22.79 | 35.44 | 34.98 | 52.09 | 79.72 | Learning to Anticipate Egocentric Actions by Imagination | - |
| ED | 8.08 | 16.07 | 29.35 | 18.19 | 38.83 | 74.49 | RED: Reinforced Encoder-Decoder Networks for Action Anticipation | |
| RULSTM [24, 23] | 14.39 | 22.78 | 33.04 | 33.73 | 50.95 | 79.55 | What Would You Expect? Anticipating Egocentric Actions with Rolling-Unrolling LSTMs and Modality Attention |
0 of 7 row(s) selected.