Co-occurring keywords · papers mentioning both
Papers
Diffuman4D: 4D Consistent Human View Synthesis from Sparse-View Videos with Spatio-Temporal Diffusion Models
ICCV 2025
UNIST: Unpaired Neural Implicit Shape Translation Network
CVPR 2022