Rethinking the compositionality of point clouds through regularization in the hyperbolic space.
Antonio MontanaroDiego ValsesiaEnrico MagliPublished in: NeurIPS (2022)
Keyphrases
- point cloud
- surface reconstruction
- underlying manifold
- structure from motion
- point cloud data
- point sets
- laser scanner
- architectural scenes
- triangular mesh
- multi view stereo
- multi dimensional scaling
- partial differential equations
- reproducing kernel hilbert space
- similarity search
- training samples
- high resolution
- training data