With the release of SDXL-Turbo in late Nov 2023, it became feasible to perform text-guided image to image generation in real-time on a consumer grade GPU. Inspired by this progress and my little cousins' artwork, I put together a web app that integrates an interactive canvas & paint tool with the capabilities of SDXL-Turbo.
The project is called Draw2Img, and I want to share it here in the hopes that you or your family enjoys it as much as we have. What makes this project unique is the combination of:
1) a simple and accessible web based UI for younger or non-technical users
2) the stunning speed & quality of image generation
3) user friendly & LAN party ready (easy to run, multiple concurrent users supported)
Thanks in advance for your feedback and support, cheers!
P.S. Here are some recent HN threads that are related and may be of interest:
StreamDiffusion: A pipeline-level solution for real-time interactive generation
https://news.ycombinator.com/item?id=38749434
SDXL Turbo: A Real-Time Text-to-Image Generation Model