Sketch-Based Text-to-Image Generation

Text-to-image tools keep improving and introducing new features every day. This research covers a new approach that makes picture generation even easier and more accurate as it takes a sketch as an input together with a text prompt.

The method doesn't require training a dedicated model or a specialized encoder for the task. Its per-pixel training "offers flexibility and locality which allows the technique to perform well on out-of-domain sketches, including free-hand style drawings." The method generates a natural highly-detailed image with a style control.

You can find more information about the project here. Also, don't forget to join our Reddit page and our Telegram channel, follow us on Instagram and Twitter, where we share breakdowns, the latest news, awesome artworks, and more.