DiHuR uses learnable SMPL-vertex tokens and a 2D diffusion prior to reconstruct detailed 3D humans from sparse multi-view images, beating prior models on standard datasets.
Mvsnerf: Fast general- izable radiance field reconstruction from multi-view stereo
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DiHuR: Diffusion-Guided Generalizable Human Reconstruction
DiHuR uses learnable SMPL-vertex tokens and a 2D diffusion prior to reconstruct detailed 3D humans from sparse multi-view images, beating prior models on standard datasets.