Listening Head Generation on ViCo
15.67FD (Expression)ECHO
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ECHORepresentation=56-d FLAME blendshape2026.03 | 15.67 | 0.04 | 0.01 | 0.07 | 3.71 | 2.93 | 2.24 | 0.02 | 0.45 | 0.01 | |
| AvatarForcingRepresentation=56-d FLAME blendshape2026.03 | 16.64 | 0.05 | 0.01 | 0.01 | 3.12 | 3 | 2.8 | 0.03 | — | — | |
| INFPRepresentation=56-d FLAME blendshape, Protocol=reproduced version reported in [23]2026.03 | 17.52 | 0.07 | 0.01 | 0.07 | 2.19 | 3.2 | 2.1 | 0.03 | 0.51 | 0.01 | |
| InterTalk2026.06 | 18.33 | 0.04 | 0.05 | 0.01 | 4.85 | 4.07 | — | — | 0.48 | 0.01 | |
| ARIGRepresentation=76-d 3DMM coefficients2026.03 | 18.39 | 0.06 | 0.05 | 0.01 | 4.82 | 3.94 | 2.91 | 0.17 | — | — | |
| ARIG2026.06 | 18.39 | 0.05 | 0.05 | 0.01 | 4.82 | 3.94 | — | — | — | — | |
| INFP2026.06 | 18.63 | 0.07 | — | — | 4.78 | 3.92 | — | — | 0.51 | 0.01 | |
| DualTalkRepresentation=56-d FLAME blendshape2026.03 | 22.27 | 0.05 | — | — | — | — | — | — | 0.58 | 0.01 | |
| DualTalk2026.06 | 22.27 | 0.04 | 0.07 | 0.03 | 4.36 | 3.84 | — | — | 0.58 | 0.01 | |
| DIMRepresentation=56-d FLAME blendshape2026.03 | 23.88 | 0.06 | 0.06 | 0.03 | 3.71 | 2.35 | 1.53 | 0.02 | 0.7 | 0.01 | |
| DIM2026.06 | 23.88 | 0.06 | 0.06 | 0.03 | 3.71 | 2.35 | — | — | 0.7 | 0.02 | |
| L2LRepresentation=56-d FLAME blendshape2026.03 | 33.93 | 0.06 | 0.06 | 0.08 | 2.77 | 2.66 | 0.83 | 0.02 | 0.93 | 0.01 | |
| RLHGRepresentation=56-d FLAME blendshape2026.03 | 39.02 | 0.07 | 0.08 | 0.02 | 3.62 | 3.17 | 1.52 | 0.02 | 0.86 | 0.01 |