Shape Representation using Gaussian Process mixture models Traditional explicit 3D representations, such as point clouds and meshes, demand significant storage to capture fine geometric details and require complex indexing systems for surface lookups, making ...

Lightweight & interpretable. No neural components. Effective under sparse sampling.
📄 arXiv: arxiv.org/abs/2604.00862
💻 Code: github.com/POSE-Lab/GP-...
🏛️ ISPRS 2026: isprs2026toronto.com
#3DVision #GaussianProcesses #ShapeRepresentation