DreamStyle: A Unified Framework for Video Stylization

Mengtian Li^✉, Jinshu Chen, Songtao Zhao^‡, Wanquan Feng, Pengqi Tu, Qian He

Intelligent Creation, ByteDance

^✉Corresponding author, ^‡Project lead

Paper Github

Gallery

Comparison with Previous Works

Text-guided Stylization

Style-image-guided Stylization

First-frame-guided Stylization

Extended Applications

Multi-style Fusion

Long-video Stylization

Ethical Considerations

The reference images and videos used in this webpage are sourced from public domains or generated by AI, and are intended solely to demonstrate the capabilities of this research. If there are any concerns, please contact us and we will delete it in time.

BibTex

If you find DreamStyle useful in your research, please kindly cite our paper:

@misc{li2026dreamstyle,
    title={DreamStyle: A Unified Framework for Video Stylization}, 
    author={Mengtian Li and Jinshu Chen and Songtao Zhao and Wanquan Feng and Pengqi Tu and Qian He},
    year={2026},
    eprint={2601.02785},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2601.02785}, 
}