Key points are not available for this paper at this time.
Recently, computer vision has focused on denoising diffusion models, which have produced amazing outcomes for generative modeling. The idea of modeling the joint probability distribution of input and output data is the basis of generative diffusion models. They generate new data using random samples at each step using an iterative process. This paper is a survey on diffusion models. It introduces the mathematical background behind diffusion models especially the denoising diffusion probabilistic models and the different categories of diffusion models.
Sallami et al. (Sat,) studied this question.