Face as an Index: Knowing Who Is Who Using a PDA
Speaker
Jie Yang
Senior System Scientist, Human-Computer Interaction Institute, Carnegie Mellon University
When
-
Where
Newell-Simon Hall 1305 (Michael Mauldin Auditorium)
Description
In this talk, I will present a PDA-based system for extending human memory or/and information retrieval using a human face as the lookup index. The system can help a user to remember names of people whom he/she has met before, and find useful information, such as names and research interests, about people whom he/she is interested in talking to. The system uses a captured face image as the lookup index to retrieve information from some available resource such as departmental directory, web sites, personal homepages, etc. I will introduce algorithms for image preprocessing to enhance the quality of the image by sharpening focus, normalizing both lighting condition and head rotation, and a unified LDA/PCA algorithm for face recognition. I will also address design issues of the interface to assist in visualization and comprehension of retrieved information. Finally I will discuss user study and experiment results to demonstrate the feasibility of the proposed system.
Speaker's Bio
Jie Yang is a Senior System Scientist in the Human-Computer Interaction Institute at Carnegie Mellon University. He joined Interactive Systems Lab in 1994, where he has been leading research efforts to develop visual tracking and recognition technologies for improving human-computer interaction and enhancing human-human communication. He has involved developments of many multimodal systems in both intelligent working spaces and mobile platforms, such as gaze-based interface, lipreading system, image-based multimodal translation agent, multimodal people ID, and automatic sign translation systems. His current projects include developing technologies to robustly track people in a dynamic environment, to understand human activities in a nurse, to support remote collaboration, understanding and supporting large scale collaboration across multiple teams and projects, and to automatic extract text from nature scenes.