Towards Automatic Video-based Whiteboard Reading

  • Authors:
  • Markus Wienecke;Gernot A. Fink;Gerhard Sagerer

  • Affiliations:
  • -;-;-

  • Venue:
  • ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

As whiteboards have become a popular tool in meetingrooms, there has been a growing interest in making use ofthe whiteboard as a user interface for human computer interaction.Therefore, systems based on electronic white-boardshave been developed in order to serve as meetingassistants for e.g. collaborative working. However, as specialpens and erasers are required, the natural interactionis restricted. In order to render this communication methodmore natural it was proposed to retain ordinary whiteboardand pens and to visually observe the writing process usinga video camera [11, 9]. In this paper a prototype systemfor automatic video-based whiteboard reading is presented.The system is designed for recognizing unconstrained handwrittentext and is further characterized by an incrementalprocessing strategy in order to facilitate recognizingportions of text as soon as they have been written on theboard. We will present the methods employed for extractingtext regions, pre-processing, feature extraction, and statisticalmodeling and recognition. Evaluation results on awriter independent unconstrained handwriting recognitiontask demonstrate the feasibility of the proposed approach.