# neka-nat/cupoch

Robotics with GPU computing

Repository: https://github.com/neka-nat/cupoch
Canonical: https://ross.abutalabs.com/products/cupoch
Language: C++
License: MIT
License Family: permissive
Topics: point-cloud, cuda, pybind11, gpu, registration, python, odometry, jetson, ros, visual-odometry, voxel, triangle-mesh, robotics, collision-detection, occupancy-grid-map, pathfinding, distance-transform, gpgpu
Last push: 2026-06-25T10:31:44+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 8, longevity 100
- inputs: {"age_days": 2507, "days_push": 69, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1061, forks 119 (observed 2026-08-28T04:03:25.896280+00:00)

## What it is
Cupoch is a C++/Python library that implements rapid 3D data processing for robotics using CUDA, based on Open3D. It provides GPU-accelerated point cloud registration, KNN search, clustering, collision detection, path planning, and visual odometry for robot systems.

## Use cases
- accelerate point cloud registration with GPU
- real-time collision avoidance for robots
- visual odometry and SLAM on Jetson
- GPU-accelerated KNN search on point clouds
- path planning with occupancy grid maps
- point cloud clustering and downsampling on GPU

## When to choose
- you need fast 3D data processing in a robotics pipeline with NVIDIA GPUs
- you want an Open3D-like API but GPU-accelerated
- you're building SLAM, tracking, or collision avoidance on Jetson or CUDA hardware
- you need Python bindings for CUDA-based point cloud algorithms

## When to avoid
- you have no NVIDIA GPU or CUDA support
- you need CPU-only processing or broad platform portability
- you need the full feature set and maturity of Open3D
- your project doesn't involve 3D data or robotics computation

## Facets
- artifact type: library
- maturity: active
- function: image-processing, computer-vision, simulation, math, sdk
- domain: robotics, computer-vision, gpu-computing, autonomous-vehicles
- platform: cpp, python, windows
- tags: point-cloud, cuda, slam, icp-registration, collision-avoidance, path-planning, odometry, ros, jetson, open3d-based, occupancy-grid, voxel-processing, pybind11, algorithms, linux, gpu, docker

## Member repositories
- neka-nat/cupoch (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:25.896280+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 2026-08-30T06:56:51.853514+00:00, confidence not recorded.
  - readme: https://github.com/neka-nat/cupoch (fetched 2026-08-28T04:03:25.896280+00:00, sha 3f271ae7cb8e)
- Data as of 2026-08-30T08:39:29.467469+00:00.