Semantic correlation promoted shape-variant context for segmentation

• Main Authors: Ding, Henghui, Jiang, Xudong, Shuai, Bing, Liu, Ai Qun, Wang, Gang
• Other Authors: School of Electrical and Electronic Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2020
• Subjects: Engineering::Computer science and engineering; Segmentation; Computer Vision
• Online Access: https://hdl.handle.net/10356/140371
• Institution: Nanyang Technological University

Context is essential for semantic segmentation. Due to the diverse shapes of objects and their complex layout in various scene images, the spatial scales and shapes of contexts for different objects have very large variation. It is thus ineffective or inefficient to aggregate various context informa...