# cleardusk/3DDFA_V2

The official PyTorch implementation of Towards Fast, Accurate and Stable 3D Dense Face Alignment, ECCV 2020.

Repository: https://github.com/cleardusk/3DDFA_V2
Canonical: https://ross.abutalabs.com/products/3ddfa_v2
Language: Python
License: MIT
License Family: permissive
Topics: eccv, 3d-face-alignment, pytorch, face-alignment, 3d-face, 3dmm, alignment, 3d, computer-vision, onnx, 3d-landmarks, single-image-reconstruction, eccv-2020
Last push: 2024-02-02T06:54:17+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2199, "days_push": 943, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3149, forks 554 (observed 2026-08-28T04:07:46.100216+00:00)

## What it is
3DDFA_V2 is the official PyTorch implementation of the ECCV 2020 paper 'Towards Fast, Accurate and Stable 3D Dense Face Alignment'. It regresses 3DMM parameters from single images to produce 3D face landmarks, dense reconstructions, pose estimation, and exports to .ply/.obj, with ONNX runtime support for fast inference.

## Use cases
- reconstruct a 3d face model from a single photo
- detect 3d facial landmarks in images
- estimate head pose from a webcam feed
- export 3d face meshes to obj or ply files
- run fast face alignment with onnx runtime
- generate uv texture maps of faces
- track facial landmarks in video

## When to choose
- you need fast, accurate 3D dense face alignment from single images
- you want 3DMM parameter regression with PyTorch or ONNX inference
- you need head pose, landmarks, or mesh export from face photos
- you want a webcam-based real-time face tracking demo

## When to avoid
- you need full-body or multi-person 3D reconstruction
- you need a general-purpose face recognition or identity verification system
- you need a maintained production service - the repo is research code with infrequent updates

## Facets
- artifact type: library
- maturity: stable
- function: computer-vision, image-processing, machine-learning, sdk
- domain: computer-vision, deep-learning, artificial-intelligence, image-processing
- platform: python, cross-platform
- tags: 3d-face-alignment, 3dmm, face-reconstruction, pytorch, onnx, eccv-2020, landmarks, head-pose-estimation, gpu

## Member repositories
- cleardusk/3DDFA_V2 (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:46.100216+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 2026-08-30T07:25:54.966408+00:00, confidence not recorded.
  - readme: https://github.com/cleardusk/3DDFA_V2 (fetched 2026-08-28T04:07:46.100216+00:00, sha 1810f1f496f4)
- Data as of 2026-08-30T08:39:29.467469+00:00.