An Annotation Engine for Supporting Video Database Population

  • Authors:

  • Marco Carrer;Leonardo Ligresti;Gulrukh Ahanger;Thomas D. C. Little

  • Affiliations:

  • Multimedia Communications Laboratory, Department of Electrical and Computer Engineering, Boston University, Boston, Massachusetts 02215, USA.;Multimedia Communications Laboratory, Department of Electrical and Computer Engineering, Boston University, Boston, Massachusetts 02215, USA.;Multimedia Communications Laboratory, Department of Electrical and Computer Engineering, Boston University, Boston, Massachusetts 02215, USA.;Multimedia Communications Laboratory, Department of Electrical and Computer Engineering, Boston University, Boston, Massachusetts 02215, USA.

  • Venue:
  • Multimedia Tools and Applications
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

Segmentation, video data modeling, and annotation areindispensable operations necessary for creating and populating avideo database. To support such video databases, annotation data canbe collected as metadata for the database and subsequently used forindexing and query evaluation. In this paper we describe the designand development of a video annotation engine, called Vane, intended to solve this problem as a domain-independent videoannotation application.Using the Vane tool, the annotation of raw video data is achievedthrough metadata collection. This process, which is performedsemi-automatically, produces tailored SGML documents whose purpose isto describe information about the video content. These documentsconstitute the metadatabase component of the video database. Thevideo data model which has been developed for the metadata, is asopen as possible for multiple domain-specific applications. The toolis currently in use to annotate a video archive comprised ofeducational and news video content.