A paper by Associate Professor Satoshi Iizuka has been accepted at ECCV 2026,
one of the most competitive international conferences in AI.
"Improving Image-to-Image Translation via a Rectified Flow Reformulation"
Satoshi Iizuka, Shun Okamoto, Kazuhiro Fukui
This paper proposes a theory that dramatically accelerates image-generation
AI models.
For more details on the proposed method, please see:
https://arxiv.org/abs/2603.20186
