Stylecodes: Encoding Stylistic Information For Image Generation
Ciara Rowles

TL;DR
This paper introduces StyleCodes, an open-source method for encoding image styles into compact 20-symbol codes, enabling easier style control in diffusion-based image generation without significant quality loss.
Contribution
It presents a novel style encoder architecture and training procedure that produces concise style codes, facilitating style sharing and control in diffusion models.
Findings
StyleCodes achieve minimal quality loss compared to traditional methods.
The approach enables style encoding into a 20-symbol base64 code.
Open-source implementation promotes accessibility and further research.
Abstract
Diffusion models excel in image generation, but controlling them remains a challenge. We focus on the problem of style-conditioned image generation. Although example images work, they are cumbersome: srefs (style-reference codes) from MidJourney solve this issue by expressing a specific image style in a short numeric code. These have seen widespread adoption throughout social media due to both their ease of sharing and the fact they allow using an image for style control, without having to post the source images themselves. However, users are not able to generate srefs from their own images, nor is the underlying training procedure public. We propose StyleCodes: an open-source and open-research style encoder architecture and training procedure to express image style as a 20-symbol base64 code. Our experiments show that our encoding results in minimal loss in quality compared to…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsHandwritten Text Recognition Techniques · Image Retrieval and Classification Techniques · Video Analysis and Summarization
MethodsFocus
