Talking Head Reconstruction on HDTF, CelebV-HQ, and CelebV-Text Samples
4.43FIDFlashLips
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| FlashLipsArchitecture=Transformer2025.12 | 4.43 | 12.31 | 74.06 | 0.687 | 0.71 | 0.86 | 32.88 | 0.94 | 0.021 | |
| FlashLipsArchitecture=UNet2025.12 | 4.75 | 15.2 | 73.81 | 0.687 | 0.7 | 0.85 | 32.86 | 0.94 | 0.022 | |
| LatentSync2025.12 | 5.3 | 36.47 | 73.1 | 0.682 | 0.55 | 0.86 | 33.61 | 0.97 | 0.015 | |
| KeySync2025.12 | 5.48 | 24.8 | 74.18 | 0.681 | 0.56 | 0.81 | 30.39 | 0.93 | 0.03 | |
| IP-LAP2025.12 | 7.91 | 39.89 | 69.57 | 0.674 | 0.4 | 0.82 | 32.97 | 0.95 | 0.033 | |
| Diff2Lip2025.12 | 8.76 | 57.96 | 68.31 | 0.672 | 0.47 | 0.79 | 25.64 | 0.94 | 0.056 | |
| TalkLip2025.12 | 13.97 | 78.66 | 64.52 | 0.667 | 0.57 | 0.79 | 32.03 | 0.94 | 0.062 | |
| DiffDub2025.12 | 16.11 | 120.24 | 57.91 | 0.673 | 0.37 | 0.58 | 26.65 | 0.89 | 0.114 |