Diff2I2P improves cross-modal registration recall to 83.0% on 7-Scenes, up from 75.8% for the 2D3D-MATR baseline, by distilling a diffusion prior into feature learning via score distillation.
Fan- tasia3d: Disentangling geometry and appearance for high- quality text-to-3d content creation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Diff$^2$I2P: Differentiable Image-to-Point Cloud Registration with Diffusion Prior
Diff2I2P improves cross-modal registration recall to 83.0% on 7-Scenes, up from 75.8% for the 2D3D-MATR baseline, by distilling a diffusion prior into feature learning via score distillation.