Extracting Semantic Video Objects

  • Authors:
  • Fuhui Long;Dagan Feng;Hanchuan Peng;Wan-Chi Siu

  • Affiliations:
  • -;-;-;-

  • Venue:
  • IEEE Computer Graphics and Applications
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

Producing large amounts of digital media data every day requires fast transmission, efficient storage, flexible manipulation, and reuse of visual content. To achieve this goal, MPEG-4 provides a content-based framework. For video, it lets users transmit, retrieve, download, store, and reuse arbitrarily shaped semantic video objects (SVOs) efficiently and interact with media sources. This article presents an SVO system that adaptively performs spatial and temporal segmentation and fusion. It also switches between user-interactive and fully automatic modes.