# minghanqin/LangSplat

Official implementation of the paper "LangSplat: 3D Language Gaussian Splatting" [CVPR2024 Highlight]

Repository: https://github.com/minghanqin/LangSplat
Canonical: https://ross.abutalabs.com/products/langsplat
Homepage: https://langsplat.github.io/
Language: Python
License: NOASSERTION
License Family: other
Topics: 3d, 3d-gaussian-splatting, 3d-reconstruction, language
Last push: 2025-10-10T09:36:25+00:00

## Health v2 (maintenance only)
Score: 47/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 46, release rhythm 35, longevity 71
- inputs: {"age_days": 996, "days_push": 327, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1077, forks 116 (observed 2026-08-28T04:03:29.576861+00:00)

## What it is
Official implementation of LangSplat, a CVPR 2024 Highlight paper that constructs a 3D language field using 3D Gaussian Splatting with CLIP-distilled language features. It enables precise, fast open-vocabulary language queries in 3D scenes, outperforming NeRF-based methods like LERF.

## Use cases
- query 3D scenes with open-vocabulary natural language
- open-vocabulary 3D object localization
- 3D semantic segmentation with language features
- render language fields faster than NeRF-based LERF
- reconstruct 3D language fields from multi-view images
- research on 3D Gaussian splatting with language grounding

## When to choose
- you need fast open-vocabulary querying of 3D scenes
- you want to reproduce or extend CVPR 2024 3D language field research
- you need 3D object localization or segmentation driven by text queries
- you want a faster alternative to LERF for 3D language fields

## When to avoid
- you need a production-ready application with a polished UI
- you lack a CUDA-capable GPU or multi-view scene captures
- you need a permissively licensed library for commercial embedding
- you only need 2D image-language retrieval without 3D reconstruction

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, image-processing, nlp, graphics, search-engine
- domain: computer-vision, artificial-intelligence, graphics
- platform: python
- tags: 3d-gaussian-splatting, 3d-language-field, open-vocabulary-queries, clip-features, sam, nerf-alternative, research-code, cvpr-2024, natural-language-processing, linux, gpu

## Member repositories
- minghanqin/LangSplat (main) score 47

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:29.576861+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 2026-08-30T06:53:16.240014+00:00, confidence not recorded.
  - readme: https://github.com/minghanqin/LangSplat (fetched 2026-08-28T04:03:29.576861+00:00, sha db8324241f81)
  - homepage: https://langsplat.github.io/ (fetched 2026-08-29T12:54:49.764669+00:00, sha 2bf709c2fa73)
- Data as of 2026-08-30T08:39:29.467469+00:00.