Speech-driven facial animation on Hybrid Audio Vocaset, LJSpeech, and FaceTalk 1.0 (test)
11.2737LSE-DFaceTalk
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| FaceTalktraining_dataset=Vocaset2023.12 | 11.2737 | 40.692 | 0.009 | 45.75 | 0.6145 | |
| Imitatortraining_dataset=Vocaset2023.12 | 11.6119 | 208.479 | 0.212 | 38.44 | 0.5384 | |
| EMOTEtraining_dataset=Vocaset2023.12 | 11.7192 | 227.924 | 0.247 | 39.44 | 0.5149 | |
| EmoTalktraining_dataset=Vocaset2023.12 | 11.7485 | 201.311 | 0.201 | 31.8 | 0.5037 | |
| CodeTalkertraining_dataset=Vocaset2023.12 | 11.8054 | 208.064 | 0.207 | 38.38 | 0.5274 | |
| FaceFormertraining_dataset=Vocaset2023.12 | 11.9848 | 215.274 | 0.222 | 38.24 | 0.5227 | |
| MeshTalktraining_dataset=Vocaset2023.12 | 12.9607 | 220.172 | 0.254 | 36.14 | 0.4855 | |
| VOCAtraining_dataset=Vocaset2023.12 | 13.6191 | 239.043 | 0.28 | 33.36 | 0.4251 |