Locally Supervised Deep Hybrid Model for Scene Recognition.

	Locally Supervised Deep Hybrid Model for Scene Recognition.
	Guo, Sheng; Huang, Weilin; Wang, Limin; Qiao, Yu
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2017
文献子类	期刊论文
英文摘要	Convolutional neural networks (CNN) have recently achieved remarkable successes in various image classification and understanding tasks. The deep features obtained at the top fully-connected layer of the CNN (FC-features) exhibit rich global semantic information and are extremely effective in image classification. On the other hand, the convolutional features in the middle layers of the CNN also contain meaningful local information, but are not fully explored for image representation. In this paper, we propose a novel Locally-Supervised Deep Hybrid Model (LS-DHM) that effectively enhances and explores the convolutional features for scene recognition. Firstly, we notice that the convolutional features capture local objects and fine structures of scene images, which yield important cues for discriminating ambiguous scenes, whereas these features are significantly eliminated in the highly-compressed FC representation. Secondly, we propose a new Local Convolutional Supervision (LCS) layer to enhance the local structure of the image by directly propagating the label information to the convolutional layers. Thirdly, we propose an efficient Fisher Convolutional Vector (FCV) that successfully rescues the orderless mid-level semantic information (e.g. objects and textures) of scene image. The FCV encodes the large-sized convolutional maps into a fixed-length mid-level representation, and is demonstrated to be strongly complementary to the high-level FC-features. Finally, both the FCV and FC-features are collaboratively employed in the LSDHM representation, which achieves outstanding performance in our experiments. It obtains 83:75% and 67:56% accuracies respectively on the heavily benchmarked MIT Indoor67 and SUN397 datasets, advancing the stat-of-the-art substantially.
URL标识	查看原文
语种	英语
内容类型	期刊论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/11561]
专题	深圳先进技术研究院_集成所
作者单位	IEEE TRANSACTIONS ON IMAGE PROCESSING
推荐引用方式 GB/T 7714	Guo, Sheng,Huang, Weilin,Wang, Limin,et al. Locally Supervised Deep Hybrid Model for Scene Recognition.[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2017.
APA	Guo, Sheng,Huang, Weilin,Wang, Limin,&Qiao, Yu.(2017).Locally Supervised Deep Hybrid Model for Scene Recognition..IEEE TRANSACTIONS ON IMAGE PROCESSING.
MLA	Guo, Sheng,et al."Locally Supervised Deep Hybrid Model for Scene Recognition.".IEEE TRANSACTIONS ON IMAGE PROCESSING (2017).