Detail-Retained Pyramid Scene Parsing Network for Semantic Segmentation

Abstract

Semantic segmentation of scene analysis is the foundation and important topic of computer vision, and it is one of the key issues of computer vision. In the macro sense, semantic segmentation is a high-level task that paves the way for scene understanding. As the core problem of computer vision, the importance of scene understanding is more and more prominent, because more and more application scenarios in reality need to infer relevant knowledge or semantics from images. These applications include autonomous driving, human-computer interaction, computational photography, image search engines, augmented reality, and more. These problems have been solved using a variety of traditional computer vision and machine learning techniques.