Audio-driven half-body human video generation on EMTD 1.0 (evaluation set) benchmark leaderboard · SOTA2 Research

Audio-driven half-body human video generation on EMTD 1.0 (evaluation set)

49.33FIDEchoMimicV2

Evaluation Results

Method	Links
EchoMimicV22024.11		49.33	598.45	0.738	21.986	2.218	7.021	7.219	0.923	25.28	0.558
EchoMimicV22024.11		49.37	599.82	0.722	21.988	2.629	6.803	6.463	0.919	25.24	0.512
EchoMimicV22024.11		49.39	593.98	0.74	21.994	2.208	7.023	7.22	0.922	25.25	0.532
EchoMimicV22024.11		49.99	602.08	0.73	21.708	2.235	6.976	7.019	0.873	23.97	0.55
EchoMimicV22024.11		50.01	605.29	0.727	21.276	2.276	6.987	7.005	0.917	25.24	0.527
EchoMimicV22024.11		50.36	603.84	0.733	21.98	2.217	7.025	7.226	0.906	24.98	0.541
EchoMimicV22024.11		50.86	602.31	0.675	19.786	2.226	7.028	7.223	0.913	25.2	0.547
EchoMimicV22024.11		51.11	620.75	0.719	21.837	2.983	6.807	6.286	0.922	25.26	0.518
EchoMimicV22024.11		51.29	610.87	0.711	20.965	2.961	6.792	6.394	0.921	25.27	0.527
EchoMimicV22024.11		51.3	620.27	0.695	20.783	2.766	6.954	7.063	0.874	22.83	0.549
Straightforward Baseline2024.11		51.53	624.98	0.664	19.269	2.779	7.208	6.706	0.825	22.57	0.508
EchoMimicV22024.11		51.61	623.56	0.689	20.038	2.904	6.807	6.985	0.896	24.37	0.541
MimicMotion2024.11		53.47	622.62	0.702	19.278	2.628	7.958	1.495	0.907	24.82	0.526
AnimateAnyone2024.11		58.98	1,016.47	0.729	20.579	3.784	13.887	0.987	0.809	23.87	0.387