Rs-dseg: Semantic Segmentation Of High-resolution Remote Sensing Images Based On A Diffusion Model Component With Unsupervised Pretraining

Diffusion models have been widely used in various fields such as natural language processing, computer vision, and time series data modeling. These models utilize gradient-based sampling and stochastic iterative denoising processes to generate images, videos, and text. They have also been applied in segmentation and classification tasks with promising results.