Text To Video Generation On Evalcrafter Text
评估指标
Motion Quality
Temporal Consistency
Text-to-Video Alignment
Total Score
Visual Quality
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Motion Quality | Temporal Consistency | Text-to-Video Alignment | Total Score | Visual Quality | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| ModelScope | 53.09 | 54.46 | 57.8 | 218 | 52.47 | VideoComposer: Compositional Video Synthesis with Motion Controllability | |
| Show-1 | 52.19 | 60.83 | 62.07 | 229 | 53.74 | Show-1: Marrying Pixel and Latent Diffusion Models for Text-to-Video Generation | |
| VideoCrafter2 | 63.98 | 61.46 | 63.16 | 243 | 54.82 | VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models | - |
| Lavie | 57.99 | 54.23 | 68.49 | 234 | 52.83 | LAVIE: High-Quality Video Generation with Cascaded Latent Diffusion Models | |
| VideoCrafter1 | 60.85 | 55.89 | 61.95 | 232 | 53.08 | VideoCrafter1: Open Diffusion Models for High-Quality Video Generation |
0 of 5 row(s) selected.