List your company

Audio-driven portrait video generation on HDTF (test)

9.489SCVASA-1
3.954125.391066.8288.26494Jun 1, 2026

Evaluation results

MethodLinks
VASA-1Resolution=512 x 512, Hardware=RTX 4090 GPU2026.069.4895.92874.2114.86337.9
Ours M=1Resolution=512 x 512, M (Number of reference images)=1, Hardware=RTX 4090 GPU2026.069.1966.02470.784.56923.3
OursM (number of reference images)=2, Resolution=512x512, Hardware=Single H100 GPU2026.069.0566.17573.949.442.3
OursM (number of reference images)=3, Resolution=512x512, Hardware=Single H100 GPU2026.068.9986.22673.943.2742.3
OursM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.068.9436.28669.962.342.3
SonicM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.068.7996.60268.943.921.7
Hallo2M (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.067.5477.81924.788.171.2
SadTalkerResolution=512 x 512, Hardware=RTX 4090 GPU2026.067.4977.4947.9193.8433.8
HalloM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.067.4577.84124.290.881.2
Hallo3M (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.067.2568.59633.776.430.27
EchoMIMIC-DistilledM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.065.5139.3534.8174.06113.3
EchoMIMICM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.065.2919.55734.1143.621.4
FantasyTalkingM (number of reference images)=1, Resolution=512x512, Hardware=Single H100 GPU2026.064.16711.14440.789.7260.36