2024
DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control
ECCV 2024poster
"Large, pretrained latent diffusion models (LDMs) have demonstrated an extraordinary ability to generate creative content, specialize to user data through few-shot fine-tuning, and condition their output on other modalities, such as semantic maps. However, are they usable as large-scale data generat…