Search papers, labs, and topics across Lattice.
1
0
2
DipGuava achieves photorealistic and personalized 3D head avatars from monocular video by explicitly disentangling facial appearance, outperforming prior methods in visual quality and quantitative performance.
