# XingangPan/DragGAN

Official Code for DragGAN (SIGGRAPH 2023)

Repository: https://github.com/XingangPan/DragGAN
Canonical: https://ross.abutalabs.com/products/draggan
Homepage: https://vcai.mpi-inf.mpg.de/projects/DragGAN/
Language: Python
License: NOASSERTION
License Family: other
Topics: artificial-intelligence, generative-adversarial-network, generative-models, image-manipulation
Last push: 2024-05-18T17:51:40+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 85
- inputs: {"age_days": 1203, "days_push": 837, "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 35755, forks 3401 (observed 2026-08-28T04:12:00.925623+00:00)

## What it is
Official PyTorch implementation of DragGAN (SIGGRAPH 2023), an interactive point-based image manipulation method built on StyleGAN3. Users drag handle points on a generated image to precisely control pose, shape, expression, and layout, with a Gradio visualizer and web demos included.

## Use cases
- drag points on a generated image to change pose or expression
- edit GAN-generated images with precise pixel control
- research interactive GAN-based image manipulation
- run a demo of point-based image editing
- manipulate layout and shape of generated photos

## When to choose
- you need the official, research-accurate implementation of DragGAN
- you want interactive point-based editing on a StyleGAN3 generative manifold
- you have a CUDA GPU or Apple Silicon for running the visualizer

## When to avoid
- you need general-purpose photo editing of arbitrary real images rather than GAN-generated ones
- you lack a GPU and need fast interactive performance
- you need a permissively licensed production-ready library (license is non-standard)

## Facets
- artifact type: library
- maturity: stable
- function: image-processing, machine-learning, deep-learning
- domain: artificial-intelligence, computer-vision, image-processing
- platform: python, windows
- tags: gan, stylegan3, image-manipulation, interactive-editing, pytorch, siggraph-2023, research-code, gradio, generative-models, gpu, linux, macos

## Member repositories
- XingangPan/DragGAN (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:00.925623+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 2026-08-29T16:26:44.710702+00:00, confidence not recorded.
  - readme: https://github.com/XingangPan/DragGAN (fetched 2026-08-28T04:12:00.925623+00:00, sha 32552d2694e1)
  - homepage: https://vcai.mpi-inf.mpg.de/projects/DragGAN/ (fetched 2026-08-29T07:47:51.758463+00:00, sha cee13fa43e48)
- Data as of 2026-08-30T08:39:29.467469+00:00.