⬢github Python · 26 ★ +23 since we first saw it · pushed 6 h ago · NOASSERTION
Neighborhood-Grid/flatpoints
Apply the principles of geometry images (https://hhoppe.com/proj/gim/) directly to point clouds. Each 3d point of the cloud become a 2d pixel, the pixels together form an image like indexing structure.
flatpoints is a Python library that flattens a 3D point cloud into a square 2D image: each point becomes one pixel, and neighboring pixels hold neighboring points on the surface. Any per-point data (coordinates, normals, colors, features) becomes an image, so 2D tools like convolutions, CNNs, SVD, and grid interpolation can be applied to point clouds.
Why now: It was just featured in a 'Show HN: 3D to 2D, point clouds flattening (geometry images)' post, showcasing a fresh application of geometry images directly to point clouds.
Who it is for: Researchers and engineers working with 3D point clouds (graphics, vision, ML) who want to apply image-processing or CNN techniques to them.
flatteninggeometry-images
Stars over our 193 snapshots: 3 to 26, since 2 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/Neighborhood-Grid/flatpoints