# PRBonn/semantic_suma

SuMa++: Efficient LiDAR-based Semantic SLAM (Chen et al IROS 2019)

Repository: https://github.com/PRBonn/semantic_suma
Canonical: https://ross.abutalabs.com/products/semantic_suma
Language: C++
License: MIT
License Family: permissive
Topics: rangenet-lib, suma, lidar, slam, semantic, 3d-lidar, suma-plus-plus
Last push: 2024-03-12T16:31:46+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2595, "days_push": 904, "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 1015, forks 207 (observed 2026-08-28T04:03:14.209222+00:00)

## What it is
SuMa++ is a C++ implementation of efficient LiDAR-based semantic SLAM that builds 3D semantic maps from laser range scans, combining the SuMa surfel-based SLAM system with RangeNet++ semantic segmentation. It was published at IROS 2019 and is developed by researchers at the University of Bonn.

## Use cases
- generate semantic 3D maps from LiDAR scans
- run SLAM on KITTI odometry datasets
- segment LiDAR point clouds into semantic classes while localizing
- research semantic SLAM for autonomous driving
- visualize 3D LiDAR maps with a Qt/OpenGL GUI
- evaluate loop closure with semantic consistency checks

## When to choose
- you need LiDAR-only SLAM with semantic labels on KITTI-format data
- you want a research-grade reference implementation of semantic SLAM
- you work with 3D LiDAR scans and want surfel-based mapping with segmentation

## When to avoid
- you need real-time SLAM on embedded hardware with limited GPU resources
- your data is not in KITTI scan format and conversion is impractical
- you need a maintained production system rather than a research codebase
- you prefer visual or RGB-D SLAM over LiDAR-based approaches

## Facets
- artifact type: application
- maturity: maintenance
- function: computer-vision, machine-learning, graphics, simulation
- domain: robotics, autonomous-vehicles, computer-vision, machine-learning
- platform: cpp
- tags: lidar, slam, semantic-mapping, 3d-perception, kitti, rangenet, point-cloud, ros, gtsam, pose-graph-optimization, linux, docker, gpu

## Member repositories
- PRBonn/semantic_suma (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:14.209222+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 2026-08-30T07:11:24.320213+00:00, confidence not recorded.
  - readme: https://github.com/PRBonn/semantic_suma (fetched 2026-08-28T04:03:14.209222+00:00, sha 9293c4975b75)
- Data as of 2026-08-30T08:39:29.467469+00:00.