naver-ai/StyleKeeper
PyTorch implementation of diffusion models for text-to-stylized image generation using novel self-attention swapping techniques.
Not currently ranked — collecting fresh signals.
star history
The repository provides official implementations of StyleKeeper and VSP, two approaches for controlling visual style in text-to-image diffusion models without requiring fine-tuning. Both methods operate during the denoising process by swapping key and value projections in self-attention layers to transfer style from reference images while preventing content leakage. Published at ICCV 2025 with VSP winning Best and Oral recognition at CVPRW AI4CC 2024.
Frequently asked
- What is naver-ai/StyleKeeper?
- PyTorch implementation of diffusion models for text-to-stylized image generation using novel self-attention swapping techniques.
- Is StyleKeeper open source?
- Yes — naver-ai/StyleKeeper is open source, released under the Apache-2.0 license.
- What language is StyleKeeper written in?
- naver-ai/StyleKeeper is primarily written in Python.
- How popular is StyleKeeper?
- naver-ai/StyleKeeper has 482 stars on GitHub.
- Where can I find StyleKeeper?
- naver-ai/StyleKeeper is on GitHub at https://github.com/naver-ai/StyleKeeper.