mkazhdan/PoissonRecon

Health v2 · maintenance only

Adoption not part of the score

1858 stars · 470 forks observed · 2026-09-22

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

An implementation of Poisson Surface Reconstruction for turning oriented point clouds into watertight 3D meshes, evolved into a general adaptive multigrid finite-element solver. It supports arbitrary dimensions, element degrees, and point-wise/integrated constraints.

Use cases

• reconstruct a 3D mesh from a scanned point cloud
• convert LiDAR or depth-sensor points into a watertight surface
• solve adaptive finite-element systems with multigrid
• generate meshes from RGB-D scan data
• process oriented point sets into implicit surfaces

When to choose

• you need high-quality watertight surfaces from oriented point clouds
• you want a proven, well-cited geometry reconstruction algorithm
• you need a flexible adaptive multigrid finite-element solver

When to avoid

• you need mesh repair of existing meshes rather than reconstruction from points
• you want a GPU-accelerated real-time solution
• you need a high-level Python or GUI workflow

Facets

library · maturity stable

graphics simulation math graphics computer-vision windows cpp cross-platform poisson-surface-reconstruction 3d-reconstruction mesh-generation multigrid-solver finite-elements point-cloud geometry-processing algorithms linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mkazhdan/PoissonRecon")