NitroFusion: High-Fidelity Single-Step Diffusion through Dynamic Adversarial Training

Dar-Yen Chen; Hmrishav Bandyopadhyay; Kai Zou; Yi-Zhe Song

doi:10.1109/cvpr52734.2025.00717

CVPR2025顶会

NitroFusion: High-Fidelity Single-Step Diffusion through Dynamic Adversarial Training

Dar-Yen Chen, Hmrishav Bandyopadhyay, Kai Zou, Yi-Zhe Song

DOI 出版方

2025年份

9顶会引用

摘要

https://chendaryen.github.io/NitroFusion.github.io Figure 1. Our one-step diffusion pipeline generates vibrant and photorealistic images with exceptional detail in a single inference step, broadening the potential for text-to-image synthesis in applications like real-time interactive systems.