Clipped video parts estimation based on transcription

IR (HANDLE) Open Access
  • Kojo, Kai
    Graduate School and Faculty of Information Science and Electrical Engineering, Kyushu University
  • Ito, Eisuke
    Research Institute for Information Technology, Kyushu University

Bibliographic Information

Other Title
  • 文字起こしに基づく動画の切抜き部分推定

Description

In recent years, a huge number of videos have been posted every day on video sharing sites such as YouTube. Summarised videos and clipped videos become populare, and they allow longer videos to be viewed in shorter formats We are interested in some methods to realize video summarization and video clipping methods using machine learning. We assume that a manually cut video and its original video are better for machine learning as training data. For the future machine learning using massive amount of video, we develop a tool that estimates clipped video parts in its original video using transcription texts. In this paper, we describe the method of clipped video parts estimation based on transcription, a prototype tool, and estimation results.

Journal

Keywords

Details 詳細情報について

  • CRID
    1050018120835343104
  • HANDLE
    2324/7170852
  • Text Lang
    ja
  • Article Type
    conference paper
  • Data Source
    • IRDB

Report a problem

Back to top