# JiahuiYu/generative_inpainting

DeepFill v1/v2 with Contextual Attention and Gated Convolution, CVPR 2018, and ICCV 2019 Oral

Repository: https://github.com/JiahuiYu/generative_inpainting
Canonical: https://ross.abutalabs.com/products/generative_inpainting
Homepage: http://jiahuiyu.com/deepfill/
Language: Python
License: NOASSERTION
License Family: other
Topics: deepfill, image-inpainting, generative-adversarial-network, attention-model, deep-neural-networks, tensorflow
Last push: 2024-06-27T08:18:38+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3143, "days_push": 797, "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 3466, forks 798 (observed 2026-08-28T04:08:05.796769+00:00)

## What it is
An open-source implementation of DeepFill v1/v2 generative image inpainting models, featuring Contextual Attention (CVPR 2018) and Gated Convolution (ICCV 2019 Oral). It provides training and testing scripts for filling in missing or masked regions of images with realistic generated content.

## Use cases
- remove objects from photos by filling in the hole
- fill in missing or masked regions of an image
- free-form image inpainting with deep learning
- reproduce DeepFill v2 gated convolution inpainting results
- train an inpainting model on my own image dataset
- restore old damaged photos automatically

## When to choose
- you need state-of-the-art free-form image inpainting based on published research
- you want to train or fine-tune an inpainting model on your own dataset
- you are researching attention-based or gated convolution inpainting architectures

## When to avoid
- you need a production-ready, actively maintained tool - the code targets old TensorFlow 1.x releases
- you need a permissively licensed solution - the CC BY-NC license forbids commercial use
- you want a simple GUI or one-click photo editor rather than a research codebase

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python
- tags: image-inpainting, deepfill, gated-convolution, contextual-attention, generative-adversarial-network, tensorflow, research-code, non-commercial-license, linux, gpu

## Member repositories
- JiahuiYu/generative_inpainting (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:05.796769+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 2026-08-29T18:36:46.806719+00:00, confidence not recorded.
  - readme: https://github.com/JiahuiYu/generative_inpainting (fetched 2026-08-28T04:08:05.796769+00:00, sha 8d8c7087a0ba)
- Data as of 2026-08-30T08:39:29.467469+00:00.