A brief survey on RGB-D semantic segmentation using deep learning

Semantic segmentation is referred to as a process of linking each pixel in an image to a class label. With this pragmatic technique, it is possible to recognize different objects in an RGB image based on the color and texture, and hence it becomes easier to evaluate. Recently, researchers could perform semantic segmentation pretty well in RGB images. However, the methods based on RGB image lack enough information to realize semantic segmentation of complex scenes. RGB-D semantic segmentation with depth information has been proved to achieve better segmentation results by a lot of experiments, but there is a lack of a comprehensive survey. In this paper, the main purpose is to offer a detailed review of RGB-D semantic segmentation according to the research progress in recent years. Specifically, recently updated RGB-D datasets will be focused on first, and problems on RGB-D semantic segmentation will be discussed. In the end, a comprehensive analysis is carried out on recent methods and their analysis of the semantic segmentation.