Video-driven Talking Head Generation (Self-Reenactment) on NeRSemble Mono benchmark leaderboard · SOTA2 Research

Video-driven Talking Head Generation (Self-Reenactment) on NeRSemble Mono

31.98PSNRMGGTalk

Evaluation Results

Method	Links
MGGTalk2025.04		31.98	0.773	51.35	0.125	0.298
Real3DPortrait2025.04		31.54	0.667	79.09	0.139	0.325
DaGAN2025.04		31.39	0.745	82.5	0.166	0.575
OTAvatar2025.04		31.23	0.695	59.42	0.158	0.331
ROME2025.04		31.07	0.712	95.27	0.142	0.42
Styleheat2025.04		30.67	0.719	77.45	0.189	0.602
Portrait4D-v22025.04		30.29	0.789	54.95	0.182	0.367