Methods for generation of photorealistic images

Abstract

Modern computer graphics [1] is one of the most important areas of information technology development. Its use covers computer games, cinematography, architectural visualization, virtual and augmented reality systems, medicine, scientific simulations, and digital design. One of the key tasks of computer graphics is to create photorealistic images that are as close as possible to the real human perception of the world around them. The realism of a digital image [ 1-7 ] is determined by the accuracy of modeling geometry, lighting, materials, textures, and physical properties of surfaces. The development of graphics processors and specialized algorithms has allowed us to move simple raster scenes to complex real-time global illumination and ray tracing systems. In modern image generation systems, neural network methods are actively used, which significantly expand the capabilities of classical rendering. The purpose of the monograph is to analyze modern methods for generating photorealistic images, study their mathematical and hardware foundations, and study the prospects for the development of digital visualization technologies.