# kzampog/cilantro

A lean C++ library for working with point cloud data

Repository: https://github.com/kzampog/cilantro
Canonical: https://ross.abutalabs.com/products/cilantro
Language: C++
License: MIT
License Family: permissive
Topics: point-cloud, icp, ransac, convex-hull, 3d-visualization, k-means, pca, segmentation, rgbd, registration, iterative-closest-point, mds, mean-shift, spectral-clustering, clustering, non-rigid-registration, convex, model-fitting, reconstruction, 3d
Last push: 2025-06-23T18:10:37+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 28, release rhythm 35, longevity 100
- inputs: {"age_days": 3409, "days_push": 436, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1135, forks 203 (observed 2026-08-28T04:03:43.356190+00:00)

## What it is
cilantro is a lean, templated C++ library for processing 3D point cloud data, offering kd-trees, normal estimation, resampling, PCA, PLY I/O, RGBD conversion, convex hulls, and clustering algorithms. It emphasizes a clean API, minimal boilerplate, and extensibility across arbitrary numerical types and dimensions.

## Use cases
- process 3d point clouds in c++
- register point clouds with icp
- segment point clouds
- compute convex hulls of point sets
- cluster point cloud data with k-means or mean-shift
- convert rgbd images to point clouds
- estimate surface normals from raw point clouds
- read and write ply point cloud files

## When to choose
- you need a fast, header-friendly C++ alternative to PCL with less boilerplate
- you need general-dimension algorithms (kd-trees, clustering, convex hulls) beyond just 3D
- you want modular, templated point cloud pipelines with Eigen integration

## When to avoid
- you need a Python-first point cloud workflow
- you require the full breadth of PCL's feature set or its ecosystem integrations
- you need GPU-accelerated point cloud processing out of the box

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, math, data-visualization, graphics
- domain: computer-vision, graphics, robotics
- platform: cpp, windows, cross-platform
- tags: point-cloud, icp, ransac, convex-hull, clustering, segmentation, registration, rgbd, 3d, kd-tree, pca, spectral-clustering, mean-shift, k-means, model-fitting, reconstruction, algorithms, linux, macos

## Member repositories
- kzampog/cilantro (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:43.356190+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:36:47.992987+00:00, confidence not recorded.
  - readme: https://github.com/kzampog/cilantro (fetched 2026-08-28T04:03:43.356190+00:00, sha 372c7527a3ac)
- Data as of 2026-08-30T08:39:29.467469+00:00.