# andyzeng/tsdf-fusion-python

Python code to fuse multiple RGB-D images into a TSDF voxel volume.

Repository: https://github.com/andyzeng/tsdf-fusion-python
Canonical: https://ross.abutalabs.com/products/tsdf-fusion-python
Homepage: http://andyzeng.github.io/
Language: Python
License: BSD-2-Clause
License Family: permissive
Topics: rgbd, tsdf, kinect-fusion, depth-camera, 3d-deep-learning, volumetric-data, cuda, 3d-reconstruction, 3d, vision, artificial-intelligence
Last push: 2023-02-18T07:39:49+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2814, "days_push": 1292, "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 1430, forks 236 (observed 2026-08-28T04:04:42.233843+00:00)

## What it is
A lightweight Python script that fuses multiple registered RGB-D images into a projective TSDF voxel volume, from which high-quality 3D surface meshes and point clouds can be extracted. It supports optional NVIDIA GPU acceleration via PyCUDA and is part of the 3DMatch toolbox.

## Use cases
- fuse multiple RGB-D images into a TSDF voxel volume
- reconstruct a 3D mesh from depth camera frames
- generate point clouds from registered color and depth images
- preprocess volumetric data for 3D deep learning
- run KinectFusion-style TSDF fusion in Python
- convert depth image sequences into 3D surface models

## When to choose
- you need a simple, scriptable TSDF fusion pipeline in Python
- you have registered RGB-D images with known camera poses
- you want GPU-accelerated volumetric fusion on an NVIDIA GPU
- you're preparing voxel volumes for 3D deep learning research

## When to avoid
- you need real-time SLAM or camera tracking built in
- you don't have depth/RGB-D data or camera poses
- you need a maintained library with active support
- you need CPU-only performance at high frame rates

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, computer-vision, graphics, gpu-computing
- domain: computer-vision, graphics, machine-learning, robotics
- platform: python
- tags: tsdf, rgbd, 3d-reconstruction, depth-camera, voxel-volume, kinect-fusion, mesh-generation, point-cloud, linux, gpu

## Member repositories
- andyzeng/tsdf-fusion-python (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:42.233843+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 2026-08-30T04:37:10.892045+00:00, confidence not recorded.
  - readme: https://github.com/andyzeng/tsdf-fusion-python (fetched 2026-08-28T04:04:42.233843+00:00, sha 6147731eb798)
  - homepage: http://andyzeng.github.io/ (fetched 2026-08-29T11:48:41.937222+00:00, sha a3cd0b5638d2)
- Data as of 2026-08-30T08:39:29.467469+00:00.