Abstract
Introduction
% Diffusion
In recent years, diffusion models have been rapidly developed and applied to various fields, such as text-to-image generation~\cite{2022-CVPR-LDM, 2022-arxiv-t2iclip, 2022-NIPS-t2iPhotorealistic,2022-ICML-GLIDE}, image-to-image translation~\cite{2021-ICCV-ILVR, 2022-SIGGRAPH-Palette}, image editing\cite{2022-ICLR-SDEdit, 2022-CVPR-Blended, 2023-CVPR-imagic}, image super-resolution~\cite{2023-TPAMI-SRRefine, 2022-NC-SRDiff}, data augmentation~\cite{2024-arxiv-LAKE-RED},image segmentation\cite{2023-arxiv-camoDiffusion, 2023-arxiv-DMforCOD, 2024-arxiv-DiffSal}, reference-guided image generation~\cite{2023-CVPR-GLIGEN, 2023-ICCV-TF-ICON, 2024-AAAI-t2iadapter, 2023-ICCV-ConditionControl, 2023-arXiv-ProSpect}, personalized image generation~\cite{2022-arXiv-OneWord,2023-NIPS-photoswap,2023-ICCV-SVDiff,2023-CVPR-multiconcept, 2023-CVPR-DreamBooth}, text-to-video genera