VFHQ
PulseAugur coverage of VFHQ — every cluster mentioning VFHQ across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
EfficientSync achieves real-time lip sync by preserving reference textures
Researchers have developed EfficientSync, a novel real-time framework for audio-driven lip synchronization in videos. Unlike previous methods that reconstruct the lower face, EfficientSync focuses on faithfully transfer…
-
SpatialAvatar-0: New method for high-quality 4D head avatar generation
Researchers have introduced SpatialAvatar-0, a novel method for generating high-quality 4D head avatars. This approach combines a feed-forward prediction model with a per-subject refinement process, utilizing a shared G…
-
SpatialAvatar-0 advances 4D head avatar generation with novel multi-stage reconstruction
Researchers have introduced SpatialAvatar-0, a novel method for generating high-quality 4D head avatars. This system combines a feed-forward prediction stage with a per-subject refinement process, utilizing a shared Gau…