inpaint-web  by lxfater

In-browser tool for inpainting and image upscaling

Created 1 year ago
5,457 stars

Top 9.3% on SourcePulse

GitHubView on GitHub
Project Summary

Inpaint-web provides a free, open-source, browser-based solution for image inpainting and upscaling. It targets users seeking efficient image editing tools directly within their web browser, leveraging advanced client-side technologies for performance.

How It Works

The tool utilizes WebGPU and WebAssembly (WASM) to perform image processing tasks directly in the browser. This approach avoids the need for server-side computation or dedicated hardware, making it accessible and efficient for a wide range of users. The core inpainting functionality appears to be based on the MI-GAN model.

Quick Start & Requirements

  • Install/Run: npm install followed by npm run start for development.
  • Prerequisites: A modern web browser with WebGPU support.
  • Demo: https://inpaintweb.lxfater.com/

Highlighted Details

  • Browser-native implementation using WebGPU and WASM.
  • Supports inpainting (image restoration) and image upscaling.
  • Frontend code is modified from cleanup.pictures.
  • Roadmap includes integration with Segment Anything and Stable Diffusion.

Maintenance & Community

The project is maintained by lxfater, who can be contacted via WeChat (please state your purpose) or Twitter for updates and discussions.

Licensing & Compatibility

The project is described as free and open-source. The specific license is not explicitly stated in the README, but its open-source nature suggests it's intended for broad use.

Limitations & Caveats

The project is still under active development, with features like Segment Anything and Stable Diffusion integration planned for future releases. Browser compatibility may vary depending on WebGPU support.

Health Check
Last Commit

11 months ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
57 stars in the last 30 days

Explore Similar Projects

Starred by Yineng Zhang Yineng Zhang(Inference Lead at SGLang; Research Scientist at Together AI), Rodrigo Nader Rodrigo Nader(Cofounder of Langflow), and
1 more.

DiffSynth-Studio by modelscope

0.9%
10k
Open-source project for diffusion model exploration
Created 1 year ago
Updated 16 hours ago
Starred by Gabriel Almeida Gabriel Almeida(Cofounder of Langflow), Alex Yu Alex Yu(Research Scientist at OpenAI; Former Cofounder of Luma AI), and
1 more.

IOPaint by Sanster

0.1%
22k
Image editing tool for inpainting and outpainting tasks
Created 3 years ago
Updated 4 months ago
Feedback? Help us improve.