Ajou University repository

AR-CNN: an attention ranking network for learning urban perception
  • Li, Zhetao ;
  • Chen, Ziwen ;
  • Zheng, Wei Shi ;
  • Oh, Sangyoon ;
  • Nguyen, Kien
Citations

SCOPUS

4

Citation Export

DC Field Value Language
dc.contributor.authorLi, Zhetao-
dc.contributor.authorChen, Ziwen-
dc.contributor.authorZheng, Wei Shi-
dc.contributor.authorOh, Sangyoon-
dc.contributor.authorNguyen, Kien-
dc.date.issued2022-01-01-
dc.identifier.urihttps://dspace.ajou.ac.kr/dev/handle/2018.oak/32462-
dc.description.abstractAn increasing number of deep learning methods is being applied to quantify the perception of urban environments, study the relationship between urban appearance and resident safety, and improve urban appearance. Most advanced methods extract image feature representations from street-level images through conventional visual computation algorithms or deep convolutional neural networks and then directly predict the results using features. Unfortunately, these methods take color and texture information together during processing. Color and texture are prime image features, and they affect human perception and judgment differently. We argue that color and texture should be operated differently; therefore, we formulate an end-to-end learning methodology to process input images according to color and texture information before inputting it into the neural network. The processed images and the original image constitute three input streams for the triad attention ranking convolutional neural network (AR-CNN) model proposed in this study. In accordance with the aspects of color and texture, an improved attention mechanism in the convolution layer is proposed. Our objective is to obtain the scores of humans on urban appearance in accordance with the prediction results computed from pairwise comparisons generated by the AR-CNN model.-
dc.description.sponsorshipThis work was supported in part by National Natural Science Foundation of China (Grant No. 62032020), in part by Hunan Science and Technology Planning Project (Grant No. 2019RS3019), in part by Hunan Provincial Natural Science Foundation of China for Distinguished Young Scholars (Grant No. 2018JJ1025), and in part by Guangzhou Research Project (Grant No. 201902010037).-
dc.language.isoeng-
dc.publisherScience Press (China)-
dc.subject.meshAttention network-
dc.subject.meshAttribute learning-
dc.subject.meshColor and textures-
dc.subject.meshColor information-
dc.subject.meshConvolutional neural network-
dc.subject.meshLearning methods-
dc.subject.meshNeural network model-
dc.subject.meshRanking network-
dc.subject.meshTexture information-
dc.subject.meshUrban perception-
dc.titleAR-CNN: an attention ranking network for learning urban perception-
dc.typeArticle-
dc.citation.titleScience China Information Sciences-
dc.citation.volume65-
dc.identifier.bibliographicCitationScience China Information Sciences, Vol.65-
dc.identifier.doi10.1007/s11432-019-2899-9-
dc.identifier.scopusid2-s2.0-85122099698-
dc.identifier.urlhttp://www.springer.com/computer/information+systems+and+applications/journal/11432-
dc.subject.keywordattention network-
dc.subject.keywordattribute learning-
dc.subject.keywordcolour and texture-
dc.subject.keywordranking network-
dc.subject.keywordurban perception-
dc.description.isoafalse-
dc.subject.subareaComputer Science (all)-
Show simple item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Oh, Sangyoon Image
Oh, Sangyoon오상윤
Department of Software and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.