Image Representations With Spatial Object-to-Object Relations for RGB-D Scene Recognition

doi:10.1109/TIP.2019.2933728

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Image Representations With Spatial Object-to-Object Relations for RGB-D Scene Recognition
	Song, Xinhang; Jiang, Shuqiang; Wang, Bohan; Chen, Chengpeng; Chen, Gongwei
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2020
卷号	29 页码:525-537
关键词	Feature extraction Object detection Image recognition Layout Data models Recurrent neural networks Scene recognition object-to-object relation sequential representations RGB-D object detection
ISSN号	1057-7149
DOI	10.1109/TIP.2019.2933728
英文摘要	Scene recognition is challenging due to the intra-class diversity and inter-class similarity. Previous works recognize scenes either with global representations or with the intermediate representations of objects. In contrast, we investigate more discriminative image representations of object-to-object relations for scene recognition, which are based on the triplets of & x003C;object, relation, object & x003E; obtained with detection techniques. Particularly, two types of representations, including co-occurring frequency of object-to-object relation (denoted as COOR) and sequential representation of object-to-object relation (denoted as SOOR), are proposed to describe objects and their relative relations in different forms. COOR is represented as the intermediate representation of co-occurring frequency of objects and their relations, with a three order tensor that can be fed to scene classifier without further embedding. SOOR is represented in a more explicit and freer form that sequentially describe image contents with local captions. And a sequence encoding model (e.g., recurrent neural network (RNN)) is implemented to encode SOOR to the features for feeding the classifiers. In order to better capture the spatial information, the proposed COOR and SOOR are adapted to RGB-D data, where a RGB-D proposal fusion method is proposed for RGB-D object detection. With the proposed approaches COOR and SOOR, we obtain the state-of-the-art results of RGB-D scene recognition on SUN RGB-D and NYUD2 datasets.
资助项目	National Natural Science Foundation of China[61532018] ; Beijing Natural Science Foundation[L182054] ; National Program for Special Support of Eminent Professionals ; National Program for Support of Top-Notch Young Professionals ; National Postdoctoral Program for Innovative Talents[BX201700255] ; China Postdoctoral Science Foundation[2018M631583]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000497434700020
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/14960]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Jiang, Shuqiang
作者单位	Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Song, Xinhang,Jiang, Shuqiang,Wang, Bohan,et al. Image Representations With Spatial Object-to-Object Relations for RGB-D Scene Recognition[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2020,29:525-537.
APA	Song, Xinhang,Jiang, Shuqiang,Wang, Bohan,Chen, Chengpeng,&Chen, Gongwei.(2020).Image Representations With Spatial Object-to-Object Relations for RGB-D Scene Recognition.IEEE TRANSACTIONS ON IMAGE PROCESSING,29,525-537.
MLA	Song, Xinhang,et al."Image Representations With Spatial Object-to-Object Relations for RGB-D Scene Recognition".IEEE TRANSACTIONS ON IMAGE PROCESSING 29(2020):525-537.