Video Prediction On Human36M
Métriques
MAE
MSE
SSIM
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | MAE | MSE | SSIM |
|---|---|---|---|
| simvp-simpler-yet-better-video-prediction-1 | 1510 | 316 | 0.904 |
| swinlstm-improving-spatiotemporal-prediction-1 | 1190 | 332 | 0.913 |
| folded-recurrent-neural-networks-for-future | 1901.1 | 497.7 | 0.771 |
| fast-fourier-inception-networks-for-occluded | 1190 | 233 | 0.912 |
| implicit-stacked-autoregressive-model-for-1 | 1120 | 126 | 0.942 |
| memory-in-memory-a-predictive-neural-network | 1782.8 | 429.9 | 0.790 |
| disentangling-physical-dynamics-from-unknown | 1620 | 369 | 0.901 |
| eidetic-3d-lstm-a-model-for-video-prediction | 1660 | 464 | 0.869 |
| predrnn-recurrent-neural-networks-for | 1895.2 | 484.1 | 0.781 |