Objective detection and scene understanding based on deep learning

Abstract

Machine vision plays a more and more important role in the industrial and medical market. In order to promote the progress of human society, it is necessary to deeply study artificial intelligence and deep network. In some important applications, such as automatic driving and medical detection, it will be applied to target detection and semantic segmentation based on deep learning. In order to better understand a scene, one is to identify the goals we care about in the scene, and the other is to identify the categories of each part of the scene class. This involves the semantic segmentation and target detection. This project makes an in-depth study on scene understanding in three dimensions: theoretical analysis, code construction and performance comparison.