Metric: L1 (higher is better)
| # | Model↕ | L1▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | I3D-50 + LSTM | 1.2 | No | Quo Vadis, Action Recognition? A New Model and t... | 2017-05-22 | Code |
| 2 | Hopper | 0.85 | No | Hopper: Multi-hop Transformer for Spatiotemporal... | 2021-03-19 | Code |
| 3 | OPNet | 0.54 | No | Learning Object Permanence from Video | 2020-03-23 | Code |
| 4 | TFCNet | 0.47 | No | TFCNet: Temporal Fully Connected Networks for St... | 2022-03-11 | - |
| 5 | Aloe | 0.44 | No | Attention over learned object embeddings enables... | 2020-12-15 | Code |
| 6 | Loci | 0.14 | No | Learning What and Where: Disentangling Location ... | 2022-05-26 | Code |