OCNet: Object Context for Semantic SegmentationJournal: IJCV
Yuhui Yuan1,3,4{}^{\textbf{1},\textbf{3},\textbf{4}} Lang Huang2{}^{\textbf{2}} Jianyuan Guo2{}^{\textbf{2}} Chao Zhang2{}^{\textbf{2}} Xilin Chen3,4{}^{\textbf{3},\textbf{4}} Jingdong Wang1{}^{\textbf{1}} Affiliation: 1{}^{\textbf{1}}Microsoft Research Asia 2{}^{\textbf{2}}Peking University 3{}^{\textbf{3}}Institute of Computing Technology, CAS 4{}^{\textbf{4}}University of Chinese Academy of Sciences 🖂
Abstract
In this paper, we address the semantic segmentation task with a new context aggregation scheme named object context, which focuses on enhancing the role of object information. Motivated by the fact that the category of each pixel is inherited from the object it belongs to, we define the object context for each pixel as the set of pixels that belong to the same category as the given pixel in the image. We use a binary relation matrix to represent the relationship between all pixels, where the value one indicates the two selected pixels belong to the same category and zero otherwise.
原文 arXiv:1809.00916;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1809.00916v4