AudioLift: Audio-Driven Single Image to
Full-Head Gaussian Avatar via Multi-view Video Generation

More Results

Ours on HDTF

23 HDTF clips at their natural length (11–43 s). The input is the first frame of each clip and the audio is the clip's own speech; nothing else from the video is used. Rows: input portrait, frontal rendering, and the same avatar sequence from a moving camera: a full turn around the head, or a sweep across the viewing range the training cameras cover. Videos autoplay muted; Audio unmutes one subject at a time.

(Homepage)  |  Page 1  |  Page 2  |  Page 3

(Homepage)  |  Page 1  |  Page 2  |  Page 3