# facebookresearch/sam-3d-body

The repository provides code for running inference with the SAM 3D Body Model (3DB), links for downloading the trained model checkpoints and datasets, and example notebooks that show how to use the model.

Repository: https://github.com/facebookresearch/sam-3d-body
Canonical: https://ross.abutalabs.com/products/sam-3d-body
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-02-19T18:16:10+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 68, release rhythm 35, longevity 28
- inputs: {"age_days": 400, "days_push": 195, "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 3461, forks 418 (observed 2026-08-28T04:08:05.682835+00:00)

## What it is
SAM 3D Body is a promptable model for single-image full-body 3D human mesh recovery (HMR), estimating body, feet, and hand pose using the Momentum Human Rig parametric mesh representation. This repository provides inference code, model checkpoints, datasets, and example notebooks for using the model.

## Use cases
- recover 3d human mesh from a single photo
- estimate full-body pose including hands and feet from images
- run promptable human mesh recovery with keypoints or masks
- download pretrained 3d body model checkpoints
- reconstruct 3d human body from in-the-wild images

## When to choose
- you need state-of-the-art single-image 3D human mesh recovery
- you want promptable inference guided by 2D keypoints or masks
- you need robust generalization to diverse in-the-wild images
- you want a parametric body model with decoupled skeleton and shape

## When to avoid
- you need multi-person video tracking rather than single-image recovery
- you require a permissive open-source license (license is custom)
- you need real-time on-device inference on constrained hardware
- you want 3D reconstruction of objects rather than human bodies

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, artificial-intelligence
- platform: python
- tags: human-mesh-recovery, 3d-pose-estimation, inference, segment-anything, meta-ai, model-checkpoints, gpu

## Member repositories
- facebookresearch/sam-3d-body (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:05.682835+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 2026-08-29T18:36:55.212161+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/sam-3d-body (fetched 2026-08-28T04:08:05.682835+00:00, sha 983e4cfd87e6)
- Data as of 2026-08-30T08:39:29.467469+00:00.