MOCHI enables registration-free training of multi-view 3D face reconstruction by enforcing topological consistency via a pseudo-linear inverse kinematic solver, using synthetic-data-trained 2D landmarks for alignment, and new pointmap/normal losses plus test-time optimization to outperform prior art
Unsuper- vised generative 3D shape learning from natural images
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence
MOCHI enables registration-free training of multi-view 3D face reconstruction by enforcing topological consistency via a pseudo-linear inverse kinematic solver, using synthetic-data-trained 2D landmarks for alignment, and new pointmap/normal losses plus test-time optimization to outperform prior art