Video Classification on UCF101 (full)
96.8AccuracyST Swin w/ LSTCL
Evaluation Results
| Method | Links | |
|---|---|---|
| ST Swin w/ LSTCLPretraining dataset=K400 (Unsup.), Evaluation Protocol=Full fine-tuning2021.06 | 96.8 | |
| pBYOLpre-train=K400, backbone=S3D-G, param=9.1M, T=32, mod=V, evaluation protocol=fine-tuning2021.04 | 96.3 | |
| pBYOLPretraining dataset=K400 (Unsup.), Evaluation Protocol=Full fine-tuning2021.06 | 96.3 | |
| pBYOLpre-train=K400, backbone=R-50, param=31.8M, T=16, mod=V, evaluation protocol=fine-tuning2021.04 | 95.5 | |
| BraVePretraining dataset=K400 (Unsup.), Evaluation Protocol=Full fine-tuning2021.06 | 95.1 | |
| pBYOLpre-train=K400, backbone=R(2+1)D-18, param=15.4M, T=32, mod=V, evaluation protocol=fine-tuning2021.04 | 94.4 | |
| pBYOLpre-train=K400, backbone=R-50, param=31.8M, T=8, mod=V, evaluation protocol=fine-tuning2021.04 | 94.2 | |
| MMVpre-train=AS+HT, backbone=S3D-G, param=9.1M, T=32, mod=V+A+T, evaluation protocol=fine-tuning2021.04 | 92.5 | |
| CVRLpre-train=K400, backbone=R-50, param=31.8M, T=32, mod=V, evaluation protocol=fine-tuning2021.04 | 92.2 | |
| CoCLRpre-train=K400, backbone=2xS3D-G, param=9.1M, T=32, mod=V, evaluation protocol=fine-tuning2021.04 | 90.6 | |
| GDTpre-train=K400, backbone=R(2+1)D-18, param=15.4M, T=32, mod=V+A, evaluation protocol=fine-tuning2021.04 | 89.3 | |
| ST SwinPretraining dataset=K400 (Superv.), Evaluation Protocol=Full fine-tuning2021.06 | 88.9 | |
| CoCLRpre-train=K400, backbone=S3D-G, param=9.1M, T=32, mod=V, evaluation protocol=fine-tuning2021.04 | 87.9 | |
| XDCpre-train=K400, backbone=R(2+1)D-18, param=15.4M, T=32, mod=V+A, evaluation protocol=fine-tuning2021.04 | 84.2 | |
| VTHCLpre-train=K400, backbone=R-50, param=31.8M, T=8, mod=V, evaluation protocol=fine-tuning2021.04 | 82.1 | |
| SpeedNetpre-train=K400, backbone=S3D-G, param=9.1M, T=64, mod=V, evaluation protocol=fine-tuning2021.04 | 81.1 | |
| ST SwinPretraining dataset=IN-1K (Superv.), Evaluation Protocol=Full fine-tuning2021.06 | 78.1 |