# facebookresearch/DeepSDF

Learning Continuous Signed Distance Functions for Shape Representation

Repository: https://github.com/facebookresearch/DeepSDF
Canonical: https://ross.abutalabs.com/products/deepsdf
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2022-06-20T09:07:22+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2739, "days_push": 1535, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1563, forks 281 (observed 2026-08-28T04:05:04.088081+00:00)

## What it is
DeepSDF is Facebook Research's official implementation of the CVPR 2019 paper on learning continuous signed distance functions for 3D shape representation. It provides Python scripts for preprocessing meshes, training latent-code-based SDF models, reconstructing shapes, and evaluating with Chamfer and Earth Mover's Distance metrics.

## Use cases
- learn continuous signed distance functions for 3D shapes
- reconstruct 3D meshes from point clouds
- train shape representation models on mesh datasets
- preprocess mesh datasets into SDF samples
- evaluate 3D reconstruction quality with Chamfer distance
- research neural implicit shape representations

## When to choose
- you need the reference implementation of DeepSDF for research
- you want to train neural implicit SDF models on mesh collections
- you need reproducible 3D shape reconstruction baselines

## When to avoid
- you need a production-ready 3D processing pipeline
- you want actively maintained code with recent updates
- you need a general-purpose 3D mesh editing tool

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, graphics, simulation
- domain: machine-learning, computer-vision, graphics, deep-learning
- platform: python
- tags: signed-distance-functions, shape-representation, 3d-reconstruction, research-code, cvpr-2019, linux, gpu

## Member repositories
- facebookresearch/DeepSDF (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:04.088081+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 2026-08-30T04:00:15.722716+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/DeepSDF (fetched 2026-08-28T04:05:04.088081+00:00, sha 2bcf2f062ef6)
- Data as of 2026-08-30T08:39:29.467469+00:00.