Automatic tourist attraction and representative icon determination for tourist map generation

Chao Hung Lin, Jyun Yuan Chen, Shun Siang Hsu, Yun Huan Chung

Research output: Contribution to journalArticlepeer-review

10 Citations (Scopus)

Abstract

Tourist maps are designed to direct tourists to tourist attractions in unfamiliar areas. A well-designed tourist map can provide tourists with sufficient and intuitive information about places of interest. Thus, providing upto- date information on places of interest and selecting their representative icons are fundamental and important in automatic generation of tourist maps. In this article, approaches for determining places of interest and for determining their representative icons are introduced. In contrast to general digital tourist maps that use text, simple shapes, or three-dimensional models, we use photos that offer abundant visual features of places of interest as icons in tourist maps. The photos are automatically extracted from a repository of photos downloaded from photo-sharing communities. Tourist attractions and their corresponding image icons are determined by means of photo voting and photo quality assessment. Qualitative analyses, including a user study and experiments in several areas with numerous tourist attractions, indicated that the proposed method can generate visually pleasant and elaborate tourist maps. In addition, the analyses indicated that the map produced by our method is better than maps generated by related methods and is comparable to handdesigned tourist maps.

Original languageEnglish
Pages (from-to)18-28
Number of pages11
JournalInformation Visualization
Volume13
Issue number1
DOIs
Publication statusPublished - 2014 Jan

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Automatic tourist attraction and representative icon determination for tourist map generation'. Together they form a unique fingerprint.

Cite this