OCNet: Object Context for Semantic Segmentation

doi:10.1007/s11263-021-01465-9

	OCNet: Object Context for Semantic Segmentation
	Yuan, Yuhui 1,3,4; Huang, Lang 2; Guo, Jianyuan 2; Zhang, Chao 2; Chen, Xilin 3,4; Wang, Jingdong 1
	2021-05-24
发表期刊	INTERNATIONAL JOURNAL OF COMPUTER VISION
ISSN	0920-5691
页码	24
摘要	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. We propose to use a dense relation matrix to serve as a surrogate for the binary relation matrix. The dense relation matrix is capable to emphasize the contribution of object information as the relation scores tend to be larger on the object pixels than the other pixels. Considering that the dense relation matrix estimation requires quadratic computation overhead and memory consumption w.r.t. the input size, we propose an efficient interlaced sparse self-attention scheme to model the dense relations between any two of all pixels via the combination of two sparse relation matrices. To capture richer context information, we further combine our interlaced sparse self-attention scheme with the conventional multi-scale context schemes including pyramid pooling (Zhao et al. 2017) and atrous spatial pyramid pooling (Chen et al. 2018). We empirically show the advantages of our approach with competitive performances on five challenging benchmarks including: Cityscapes, ADE20K, LIP, PASCAL-Context and COCO-Stuff.
关键词	Semantic segmentation Context Self-attention
DOI	10.1007/s11263-021-01465-9
收录类别	SCI
语种	英语
资助项目	National Nature Science Foundation of China[62071013] ; National Nature Science Foundation of China[61671027] ; National Key R&D Program of China[2018AAA0100300]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000653602100003
出版者	SPRINGER
引用统计	被引频次：149[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/17549
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Yuan, Yuhui
作者单位	1.Microsoft Res Asia, Beijing, Peoples R China 2.Peking Univ, Beijing, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 4.Univ Chinese Acad Sci, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Yuan, Yuhui,Huang, Lang,Guo, Jianyuan,et al. OCNet: Object Context for Semantic Segmentation[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2021:24.
APA	Yuan, Yuhui,Huang, Lang,Guo, Jianyuan,Zhang, Chao,Chen, Xilin,&Wang, Jingdong.(2021).OCNet: Object Context for Semantic Segmentation.INTERNATIONAL JOURNAL OF COMPUTER VISION,24.
MLA	Yuan, Yuhui,et al."OCNet: Object Context for Semantic Segmentation".INTERNATIONAL JOURNAL OF COMPUTER VISION (2021):24.