Action Classification on UAV-Human CS1
44.4CS1 AccuracyT-MOR
Evaluation results
| Method | Links | |
|---|---|---|
| T-MORPre-train=PoseCap-1M (V+M+T), Backbone=UNIK, Evaluation Protocol=Fine-tuning2026.06 | 44.4 | |
| ViAPre-train=PoseCap-1M, Backbone=UNIK, Evaluation Protocol=Fine-tuning2026.06 | 43.3 | |
| ViAPre-train=Posetics, Backbone=UNIK, Evaluation Protocol=Fine-tuning2026.06 | 42.6 | |
| AimCLRPre-train=Posetics, Backbone=AGCN, Evaluation Protocol=Fine-tuning2026.06 | 40.5 | |
| T-MORPre-train=PoseCap-1M (V+M), Backbone=UNIK, Evaluation Protocol=Fine-tuning2026.06 | 40.4 | |
| 3s-CrosCLRPre-train=Posetics, Backbone=ST-GCN, Evaluation Protocol=Fine-tuning2026.06 | 40 | |
| T-MORPre-train=PoseCap-1M (V+M+T), Backbone=UNIK, Evaluation Protocol=Linear Probing2026.06 | 33.5 | |
| ViAPre-train=Posetics, Backbone=UNIK, Evaluation Protocol=Linear Probing2026.06 | 32.9 | |
| T-MORPre-train=PoseCap-1M (V+M), Backbone=UNIK, Evaluation Protocol=Linear Probing2026.06 | 27.6 | |
| UNIK(Backbone)Pre-train=None, Backbone=UNIK, Evaluation Protocol=Linear Probing2026.06 | 3.8 |