# junyanz/iGAN

Interactive Image Generation via Generative Adversarial Networks

Repository: https://github.com/junyanz/iGAN
Canonical: https://ross.abutalabs.com/products/igan
Language: Python
License: MIT
License Family: permissive
Topics: generative-adversarial-network, image-manipulation, computer-graphics, computer-vision, gan, pix2pix, dcgan, deep-learning
Last push: 2020-08-05T01:05:44+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3633, "days_push": 2220, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4005, forks 584 (observed 2026-08-28T04:08:32.153549+00:00)

## What it is
iGAN is a research application implementing interactive image generation with generative adversarial networks, letting users draw strokes that are converted into photo-realistic images in real time. It also serves as a visual debugging tool for exploring what deep generative models can produce.

## Use cases
- generate photo-realistic images from a few brush strokes
- interactively edit images constrained to the natural image manifold
- visually debug and explore what a GAN has learned
- reproduce the ECCV 2016 iGAN paper experiments
- demo interactive GAN interfaces for teaching or research

## When to choose
- you need the reference implementation of the iGAN paper
- you want an interactive drawing interface backed by a DCGAN model
- you are studying or visualizing deep generative models on small image datasets

## When to avoid
- you need a maintained, production-ready image editing tool
- you want modern PyTorch or GPU-stack support (it uses Python 2, Theano, and PyQt4)
- you need high-resolution or general-purpose image synthesis

## Facets
- artifact type: application
- maturity: abandoned
- function: machine-learning, deep-learning, image-processing, gui
- domain: computer-vision, image-processing, deep-learning, graphics, artificial-intelligence
- platform: python
- tags: gan, dcgan, interactive-image-generation, theano, pyqt, research-code, linux, desktop

## Member repositories
- junyanz/iGAN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:32.153549+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 2026-08-29T18:24:04.959585+00:00, confidence not recorded.
  - readme: https://github.com/junyanz/iGAN (fetched 2026-08-28T04:08:32.153549+00:00, sha 14f392bda997)
- Data as of 2026-08-30T08:39:29.467469+00:00.