.png){kind=icon}
.png){kind=icon}
Return
A unified gradient-flow-based GAN framework with diffusion condition guided
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2025.112083.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Our UGGAN incorporates a diffusion process into various gradient-flow-based GANs. • We provide a theoretical basis for the convergence of UGGAN. • Our Self-supervised conditions significantly influence the diversity. • We use early stop regularization to explain the diffusion conditions. • UGGAN has significantly improved over the original gradient-flow-based GAN.
Keywords:
UGGAN
diffusion process
gradient-flow-based GAN
convergence theory
early stop regularization
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
View more organizations