A nonlinear anisotropic diffusion model with forward-backward diffusivities for image denoising

Abstract

We investigate the task of adapting image generative models to different datasets without finetuneing. To this end, we introduce Semantica, an image-conditioned diffusion model capable of generating images based on the semantics of a conditioning image. Semantica is trained exclusively on web-scale image pairs, that is it receives a random image from a webpage as conditional input and models another random image from the same webpage. Our experiments highlight the expressivity of pretrained image encoders and necessity of semantic-based data filtering in achieving high-quality image generation. Once trained, it can adaptively generate new images from a dataset by simply using images from that dataset as input. We study the transfer properties of Semantica on ImageNet, LSUN Churches, LSUN Bedroom and SUN-397.

A New Fractional-order Derivative-based Nonlinear Anisotropic Diffusion Model for Biomedical Imaging

“…The weighted and well balanced based anisotropic diffusion model is given by (Prasath and Vorotnikov 2014). The smooth Gaussian kernel based diffusion model for image restoration is proposed by (Kumar and Ahmad 2014;Kumar et al 2016). Accordingly, a fractional derivative-based nonlinear anisotropic diffusion model for biomedical imaging has been presented to reduce additive Gaussian white noise in this study.…”