MIKE CHRISTEL
Senior Systems Scientist, Computer Science and Human Computer Interaction
www

I enjoy building systems and conducting research at the intersection of human-computer interaction, multimedia processing, information visualization, and digital libraries.  I have worked with digital video since 1987, and have worked on interface development and evaluation for Carnegie Mellon's Informedia digital video understanding research group since its beginning in 1994.  Informedia research makes use of speech, image and natural language processing coupled with machine learning and interface design to enable efficient access to relevant video content from large multi-terabyte digital video collections.

My work includes designing and building video surrogate interfaces like thumbnails, storyboards and skims from automatically generated metadata, as well as utilizing information visualization to support exploratory search across very large amounts of video from heterogeneous distributed sources. I am an active participant in the NIST TRECVID video retrieval evaluation forum to benchmark progress in the field.  I wish to enhance access to video libraries so that users can explore meaningful, manipulable overviews of video document sets, issue true multimodal queries, and be aided by automatic, adaptive summarizations. 

I am a member of CMU's Human-Computer Interaction Institute, and use HCI methods to investigate the effectiveness of surrogates and information visualization schemes for video. I believe that interfaces acting as dynamic, interactive summaries across video shots and documents, with support for intelligent browsing, will allow more efficient, effective use of voluminous video resources. I am also interested in automated video content extraction, the educational use of technology, the use of video for synthetic interviews, and the application of Informedia technology to the domains of education and health care.