# luigifreda/pyslam

pySLAM is a hybrid Python/C++ Visual SLAM pipeline supporting monocular, stereo, and RGB-D cameras. It provides a broad set of modern local and global feature extractors, multiple loop-closure strategies, a volumetric reconstruction module, integrated depth-prediction models, and semantic segmentation capabilities for enhanced scene understanding.

Repository: https://github.com/luigifreda/pyslam
Canonical: https://ross.abutalabs.com/products/pyslam
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: 3d-reconstruction, depth-prediction, feature-matching, global-features, local-features, rgbd-slam, scene-understanding, semantic-mapping, semantic-segmentation, semantic-understanding, slam, stereo-slam, end-to-end-reconstruction, depth-estimation, instance-segmentation, visual-odometry, volumetric-reconstruction, loop-closure, place-recognition, gaussian-splatting
Last push: 2026-08-23T07:25:36+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 2706, "days_push": 10, "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 3401, forks 542 (observed 2026-08-28T04:08:02.885853+00:00)

## What it is
pySLAM is a hybrid Python/C++ Visual SLAM pipeline supporting monocular, stereo, and RGB-D cameras with a wide range of local and global feature extractors, loop-closure methods, and volumetric reconstruction including Gaussian Splatting. It integrates modern depth prediction and semantic segmentation models for enhanced scene understanding and serves as a flexible experimental framework for VO/SLAM research.

## Use cases
- run visual SLAM with a monocular, stereo, or RGB-D camera
- compare classical and learned local feature extractors for visual odometry
- perform loop closure and place recognition with NetVLAD or CosPlace global descriptors
- build dense 3D reconstructions from depth images using TSDF volumetric integration
- estimate depth from monocular images with DepthAnything or DepthPro inside a SLAM pipeline
- segment scenes semantically while mapping with SLAM
- reconstruct 3D scenes from multiple views with DUSt3R or VGGT
- experiment with SLAM backends using g2o or GTSAM

## When to choose
- you need a flexible Python-based Visual SLAM baseline for research and experimentation
- you want to swap and benchmark different local features, global descriptors, or loop-closing methods in one framework
- you need integrated depth prediction, semantic segmentation, or volumetric reconstruction alongside SLAM
- you want interoperable Python and C++ SLAM cores with performance-versus-flexibility tradeoffs

## When to avoid
- you need a production-grade, hard-real-time SLAM system for deployment on embedded hardware
- you require a lightweight solution without heavy deep-learning model dependencies
- you need a fully C++ SLAM implementation without Python involvement
- your project cannot comply with the GPL-3.0 license

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, machine-learning, image-processing, simulation, graphics
- domain: robotics, computer-vision, autonomous-vehicles, machine-learning
- platform: python, cpp
- tags: slam, visual-odometry, rgbd, stereo, monocular, loop-closure, depth-estimation, semantic-segmentation, gaussian-splatting, tsdf, volumetric-reconstruction, place-recognition, feature-matching, pybind11, g2o, gtsam, 3d-reconstruction, linux, macos

## Member repositories
- luigifreda/pyslam (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:02.885853+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 2026-08-29T18:38:41.822597+00:00, confidence not recorded.
  - readme: https://github.com/luigifreda/pyslam (fetched 2026-08-28T04:08:02.885853+00:00, sha 1d69bd7c8cd1)
- Data as of 2026-08-30T08:39:29.467469+00:00.