Image-to-Video Generation on DL3DV-10K 1K (test)
15.23PSNRSFT
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SFTBase Model=CogVideoX-I2V-5B2026.01 | 15.23 | 0.509 | 0.639 | 0.982 | 0.665 | 0.628 | 44.67 | 33 | 52.67 | 35 | |
| VideoGPABase Model=CogVideoX-I2V-5B2026.01 | 15.19 | 0.51 | 0.608 | 0.986 | 0.638 | 0.564 | 74 | 56 | 57.67 | 76 | |
| Epipolar-DPOBase Model=CogVideoX-I2V-5B2026.01 | 15.07 | 0.479 | 0.615 | 0.984 | 0.646 | 0.571 | 67.33 | 51.33 | 56.67 | 66 | |
| Baseline-I2VBase Model=CogVideoX-I2V-5B2026.01 | 14.57 | 0.455 | 0.653 | 0.976 | 0.687 | 0.706 | — | — | — | — |