• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Inventive Computation Technologies >3D CBVR: An efficient video retrieval using MapReduce framework with TB-PCT
【24h】

3D CBVR: An efficient video retrieval using MapReduce framework with TB-PCT

机译：3D CBVR：使用带有MapReduce框架和TB-PCT的高效视频检索

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Upgrowth of various technologies brings up the key features for effective storage and retrieval of digital video content. Such significant methodologies have to be sculpted with perfection to give accurate results. This paper proposes MapReduce framework which is molded for 3D Content Based Video Retrieval (CBVR) application along with key frame extraction, feature extraction, building up Bag of Visual Words (BoVW) and matching. We initiate with the process of key frame extraction, followed by shape feature extraction based on combining topology and geometry. Further code book generation is held based on Threshold Based - Predictive Clustering Tree (TB-PCT) algorithm for construction of BoVW is performed by Numeric Semantic Analysis (i.e.) error free. On the whole matching performance is carried out by the combination of soft-weighting scheme with L2 distance function. Our final experimental results show better accuracy and positive results than the state-of-the-art algorithms.
机译：各种技术的兴起带来了有效存储和检索数字视频内容的关键功能。这样的重要方法必须完美地雕刻才能给出准确的结果。本文提出了MapReduce框架，该框架可用于3D基于内容的视频检索（CBVR）应用程序，以及关键帧提取，特征提取，构建视觉单词袋（BoVW）和匹配功能。我们先从关键帧提取过程开始，然后再基于结合拓扑和几何的形状特征提取。基于基于阈值的预测聚类树（TB-PCT）算法来构建BoVW，可通过数字语义分析（即无错误）执行进一步的代码簿生成。总体上，通过将软加权方案与L2距离函数相结合，可以实现匹配性能。我们的最终实验结果显示出比最新算法更好的准确性和积极的结果。

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