Differentiation and Decoding of the Spatial Modulations of Textures by the Multilayer Convolutional Neural Networks

Author:

Yavna Denis V.1,Babenko Vitaly V.1,Stoletniy Alexander S.1,Shchetinina Daria P.1,Alekseeva Dariya S.1

Affiliation:

1. Southern Federal University

Abstract

The paper constitutes a short review of the second-order visual mechanisms studies. Their contribution to the process of the visual attention controlling is being of great interest today. Basic and neural network approaches in the modeling of the second-order visual mechanisms are discussed. The authors report the results of network training when modulated textures were used as training sets, and also present, as an example, the architecture of fast-learning classifier with accuracy more than 98% on test set. The representations obtained through learning are demonstrated. The results of convolutional autoencoders’ training to extract the envelope of the textures, that are modulated in contrast, orientation, and spatial frequency, are presented as well. The successful learning architectures are given as examples. The authors assume that using of convolutional networks in the modeling of the second-order visual mechanisms provides the great perspective, while the results can be used in the algorithms of saliency maps development.

Funder

Российский Фонд Фундаментальных Исследований

Publisher

Russian Foundation for Basic Research

Reference20 articles.

1. J.D. Victor, M.M. Conte, C.F. Chubb Annu. Rev. Vis. Sci., 2017, 3, 275. DOI: 10.1146/annurev-vision-102016-061316.

2. A. Barbot, M.S. Landy, M. Carrasco J. Vis., 2012, 12(8). DOI: 10.1167/12/8/6.

3. V.V. Babenko, P.N. Ermakov Vision and the Binding Problem [Zrenie i problema svyazyvaniya], RF, Moscow, Credo Publ., 2013, 159 pp. (in Russian).

4. V.M. Bondarko, M.V. Danilova, N.N. Krasilnikov, L.I. Leushina, A.A. Nevskaya, J.E. Shelepin Spatial vision [Prostranstvennoe zrenie], RF, Saint Petersburg, Nauka Publ., 1999, 211 pp. (in Russian).

5. N.V. Graham Vision Res., 2011, 51(13), 1397. DOI: 10.1016/j.visres.2011.02.007.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3