# facebookresearch/sam-3d-objects

SAM 3D Objects

Repository: https://github.com/facebookresearch/sam-3d-objects
Canonical: https://ross.abutalabs.com/products/sam-3d-objects
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-06-02T21:16:44+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 35, longevity 24
- inputs: {"age_days": 338, "days_push": 92, "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 7322, forks 874 (observed 2026-08-28T04:09:58.220021+00:00)

## What it is
SAM 3D Objects is a foundation model from Meta that reconstructs full 3D shape geometry, texture, and layout from a single image, with code, model weights, and a benchmark released. It is designed for real-world scenes with occlusion and clutter, using progressive training and a human-feedback data engine.

## Use cases
- convert a photo of an object into a 3D model
- reconstruct 3D geometry and texture from a single image
- generate 3D assets for games or AR from photos
- benchmark 3D object reconstruction models
- extract 3D shapes from cluttered real-world scenes

## When to choose
- you need single-image to 3D reconstruction with geometry and texture
- you want a state-of-the-art foundation model for 3D object generation
- your images contain occlusion and clutter typical of real-world scenes
- you want to evaluate against a challenging 3D reconstruction benchmark

## When to avoid
- you need human body or pose reconstruction (use SAM 3D Body instead)
- you need multi-view or video-based 3D reconstruction
- you lack GPU resources for running large vision models
- you need a permissively licensed model for commercial use without checking the custom license

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, computer-vision, graphics, sdk
- domain: computer-vision, deep-learning, artificial-intelligence, graphics, image-processing
- platform: python, cross-platform
- tags: 3d-reconstruction, single-image-to-3d, foundation-model, mesh-generation, texture-generation, segment-anything, meta-ai, checkpoints, benchmark, gpu, linux

## Member repositories
- facebookresearch/sam-3d-objects (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:58.220021+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 2026-08-29T17:38:39.431494+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/sam-3d-objects (fetched 2026-08-28T04:09:58.220021+00:00, sha 3fe425d30173)
- Data as of 2026-08-30T08:39:29.467469+00:00.