首页> 外文会议>IEEE Conference on Multimedia Information Processing and Retrieval >A Large-Scale Analysis of Regional Tendency of Twitter Photos Using Only Image Features
【24h】

A Large-Scale Analysis of Regional Tendency of Twitter Photos Using Only Image Features

机译:仅使用图像特征对Twitter照片的区域趋势进行大规模分析

获取原文

摘要

It is common activities for many people to post photos on open SNSs such as Twitter and Instagram where anyone can see all the posted photos. The difference on idea of the privacy issue at SNSs depends on culture and history. In fact, the people in Eastern Asian such Japan and Korea do not like to post human face photos to SNSs, while Western and South-Eastern-Asian people posts many human face photos. To make this fact clear and explore it deeper, we analyzed geotagged photos collected from the Twitter streams by classifying them into five kinds of rough categories, "people", "building", "document", "scene" and "food". As results, we found there are quite large differences on regional tendency of posted Twitter photos regarding the photo genre distributions.
机译:对于许多人来说,很多人在开放的SNS上发布照片,如Twitter和Instagram,有人可以看到所有发布的照片​​。 SNSS在SNS的隐私问题的思想差异取决于文化和历史。事实上,东亚的人民这样的日本和韩国不喜欢将人类脸部照片发布给SNSS,而西南部和东南亚人民张贴了许多人类脸照片。为了使这一事实清除并探索它更深,我们通过将它们分为五种粗略类别,“人”,“建筑”,“文件”,“场景”和“食物”来分析从Twitter流中收集的地理标记照片。作为结果,我们发现关于Photo Proite分布的Twitter照片的区域趋势存在相当大的差异。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号