Speech-driven talking face generation on Wild Dataset
8.89LSE-DHRTF
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| HRTF2025.10 | 8.89 | 9.23 | 0.14 | 0.75 | 29.68 | 28.76 | 0.0035 | |
| HRTF-SCFP2025.10 | 9.94 | 8.68 | 0.18 | 0.73 | 29.21 | 29.19 | 0.0038 | |
| VideoRetalkingVenue=SIGGRAPH 20222025.10 | 12.56 | 4.21 | 0.17 | 0.63 | 29.51 | 41.19 | 0.0049 | |
| SyncTalkVenue=CVPR 20242025.10 | 13.55 | 4.55 | 0.39 | 0.68 | 28.82 | 56.47 | 0.0063 | |
| AniportaitVenue=ECCV 20242025.10 | 14.64 | 4.42 | 0.45 | 0.63 | 28.14 | 54.85 | 0.0045 | |
| HalloVenue=arXiv 2024.72025.10 | 18.57 | 5.9 | 0.33 | 0.64 | 28.67 | 46.59 | 0.0062 | |
| Real3D-portraitVenue=ICLR 20242025.10 | 19.07 | 3.27 | 0.56 | 0.64 | 28.47 | 48.38 | 0.0056 | |
| AnitalkerVenue=ACM MM 20242025.10 | 19.5 | 4.26 | 0.36 | 0.68 | 28.73 | 48.96 | 0.0041 |