Video Frame Prediction on Kinetics-600
3.3gFVDW.A.L.T-L
Evaluation results
| Method | Links | |
|---|---|---|
| W.A.L.T-LGenerator #Params=313M2026.03 | 3.3 | |
| EVATokTokenizer #Params=145M, Generator #Params=633M2026.03 | 4 | |
| MAGVIT-v2-MLMGenerator Parameters=307M, Number of Tokens=12802024.10 | 4.3 | |
| MAGVIT-v2-MLMGenerator #Params=307M2026.03 | 4.3 | |
| EVATokTokenizer #Params=145M, Generator #Params=327M2026.03 | 4.6 | |
| LARP-L-LongTokenizer Parameters=173M, Generator Parameters=632M, Number of Tokens=1024, Training Strategy=More epochs2024.10 | 5.1 | |
| LARP-L-LongTokenizer #Params=173M, Generator #Params=632M2026.03 | 5.1 | |
| LARP-LTokenizer Parameters=173M, Generator Parameters=343M, Number of Tokens=10242024.10 | 6.2 | |
| LARP-L-LongTokenizer Parameters=173M, Generator Parameters=343M, Number of Tokens=1024, Training Strategy=More epochs2024.10 | 6.2 | |
| LARP-L-LongTokenizer #Params=173M, Generator #Params=343M2026.03 | 6.2 | |
| MAGVIT-MLMTokenizer Parameters=158M, Generator Parameters=306M, Number of Tokens=10242024.10 | 9.9 | |
| MAGVIT-L-FPPre-trained on additional large video data=false2022.12 | 9.9 | |
| MAGVIT-L-FPModel Size=464M2022.12 | 9.9 | |
| MAGVIT-MLMTokenizer #Params=158M, Generator #Params=306M2026.03 | 9.9 | |
| AdapTokTokenizer #Params=195M, Generator #Params=633M2026.03 | 11 | |
| Video DiffusionPre-trained on additional large video data=false2022.12 | 16.2 | |
| Video Diffusion2022.12 | 16.2 | |
| RaMViDPre-trained on additional large video data=false2022.12 | 16.5 | |
| RaMViD2022.12 | 16.5 | |
| MAGVIT-B-FPPre-trained on additional large video data=false2022.12 | 24.5 | |
| MAGVIT-B-FPModel Size=128M2022.12 | 24.5 | |
| TransframerPre-trained on additional large video data=false2022.12 | 25.4 | |
| Transframer2022.12 | 25.4 | |
| TrIVD-GAN-FPPre-trained on additional large video data=false2022.12 | 25.7 | |
| TrIVD-GAN-FP2022.12 | 25.7 | |
| OmniTokenizerTokenizer Parameters=82.2M, Generator Parameters=650M, Number of Tokens=12802024.10 | 32.9 | |
| OmniTokenizerTokenizer #Params=82.2M, Generator #Params=650M2026.03 | 32.9 | |
| PhenakiPre-trained on additional large video data=false2022.12 | 36.4 | |
| Phenaki2022.12 | 36.4 | |
| CCVSPre-trained on additional large video data=false2022.12 | 55 | |
| CCVS2022.12 | 55 | |
| DVD-GAN-FPPre-trained on additional large video data=false2022.12 | 69.1 | |
| CogVideoGenerator Parameters=9.4B, Number of Tokens=20652024.10 | 109.2 | |
| CogVideoPre-trained on additional large video data=true2022.12 | 109.2 | |
| CogVideo2022.12 | 109.2 | |
| CogVideoGenerator #Params=9.4B2026.03 | 109.2 | |
| Video TransformerPre-trained on additional large video data=false2022.12 | 170 | |
| LVTPre-trained on additional large video data=false2022.12 | 224.7 |