Text-to-Video Generation on WebVid (test)
61.52FIDAKiRa
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| AKiRaBackbone=AnimateDiff [18], Camera control=AKiRa (ours)2024.12 | 61.52 | 332.3 | 328.7 | 0.438 | 1.48 | 70.97 | 96.9 | 97.4 | 95.7 | |
| CameraCtrlBackbone=AnimateDiff [18], Camera control=CameraCtrl [21]2024.12 | 61.62 | 384.5 | 355.2 | 0.73 | 1.807 | 53.74 | 96.5 | 96 | 94.2 | |
| AnimateDiffBackbone=AnimateDiff [18], Camera control=None2024.12 | 65.09 | 835.2 | 768.6 | 4.354 | 3.048 | 0 | 84.5 | 84.3 | 79.2 | |
| MotionCtrlBackbone=AnimateDiff [18], Camera control=MotionCtrl [56]2024.12 | 104.64 | 822.9 | 449.7 | 0.443 | 1.857 | 58.5 | 96.1 | 96.5 | 93.9 |