stereomosaicswithslantparallelprojectionsfrommany.docVIP

Automated Registration of High Resolution Images from Slide Presentation and Whiteboard Handwriting via a Video Camera Weihong Li Dept. of Computer Science Graduate Center City University of New York New York, NY 10016 wli@ Hao Tang Dept. of Computer Science City College City University of New York New York, NY 10031 ccnytang@ Zhigang Zhu Dept. of Computer Science City College Graduate Center City University of New York New York, NY 10031 zhu@ Abstract A PowerPoint?(PPT) slide presentation and a whiteboard handwriting capture system, when used together, could provide a better means to present a lecture in a classroom and later to deliver more effective e-lectures for off-campus students. When the PPT slides are projected on the whiteboard where the instructor annotates, modifies, or expands the PPT presentation, geometrically registering high-resolution images from the two sources yields high quality digital presentations. Since the two sources do not share the same materials, one with printing notes and the other with handwriting notes, we use a low-cost digital camera as a bridge to align the two sources. We present a hybrid registration approach to align the PPT and handwriting images. We use domain knowledge in a classroom presentation (i.e., the illumination of the slide projection, the timing information for both slide changes and handwriting page creation) to align the camera views with the digital slides and to extract handwriting contents from clustered video scenes. We propose a coarse-to-fine content matching method to register handwriting contents captured by the video camera and the whiteboard capture system. Experimental results are presented to validate our approach. 1. Introduction A domain of primary importance in the future of web-based technology and digital libraries is distance and electronic education (e-learning). Multimedia materials from classrooms and seminars are rich sources of information. Todays first generation e