Vehicle Logo Recognition and attributes prediction by multi-task learning with CNN

Yizhang Xia, Jing Feng, Bailing Zhang

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

25 Citations (Scopus)

Abstract

Vehicle Logo Recognition(VLR) has been an important study field in intelligent Transportation system (ITS). This paper proposes to recognize vehicle logo and predict logo attributes by combining Convolutional Neural Network (CNN) with Multi-Task Learning(MTL). In order to accelerate convergence of multi-task model, an adaptive weight training strategy is employed. To verify the algorithm, the Xiamen University Vehicle logo recognition dataset is extended into a larger vehicle logo dataset including 15 brands, 6 visual attributes and 3 no-visual attributes. The experiment results indicate that the proposed multi-task CNN model perform well for both of logo classification and attribution prediction with overall accuracy 98.14%.

Original languageEnglish
Title of host publication2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
EditorsJiayi Du, Chubo Liu, Kenli Li, Lipo Wang, Zhao Tong, Maozhen Li, Ning Xiong
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages668-672
Number of pages5
ISBN (Electronic)9781509040933
DOIs
Publication statusPublished - 19 Oct 2016
Event12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016 - Changsha, China
Duration: 13 Aug 201615 Aug 2016

Publication series

Name2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016

Conference

Conference12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
Country/TerritoryChina
CityChangsha
Period13/08/1615/08/16

Keywords

  • Con-volutional Neural Networks(CNNs)
  • Multi-Task Learning(MTL)
  • Vehicle Logo Recognition (VLR)

Fingerprint

Dive into the research topics of 'Vehicle Logo Recognition and attributes prediction by multi-task learning with CNN'. Together they form a unique fingerprint.

Cite this