William Mong Distinguished Lecture in Engineering and Computer Science entitled ‘Prior, Context and Interactive Computer Vision’ to be delivered by Dr. Harry Shum Managing Director, Microsoft Research Asia (Beijing, China).
Details of the Talk
Dr. Harry Shum Managing Director, Microsoft Research Asia (Beijing, China)
May 15, 2006 (Monday)
Lecture Theatre A,
Chow Yei Ching Building, HKU
For many years, computer vision researchers have worked hard chasing the illusive goals such as “Can the robot find a boy in the scene?” or “Can your vision system automatically segment the cat from the background?”. These tasks require a lot of prior knowledge and contextual information. How to incorporate prior knowledge and contextual information into vision systems is very challenging. In his talk, Dr. Shum will share the proposition that many difficult vision tasks can only be solved with interactive vision systems, by combining powerful and real-time vision techniques with intuitive and clever user interfaces. Dr. Shum will show two interactive vision systems that his team has developed recently, namely Lazy Snapping (Siggraph 2004) and Image Completion (Siggraph 2005). While Lazy Snapping cuts out an object from a picture using graph cut, Image Completion recovers unknown region in a picture with belief propagation. A key element in designing such interactive systems is how to model the user’s intention using conditional probability (context) and likelihood associated with user interactions. Given how ill-posed most image understanding problems are, it is proposed that interactive computer vision should be the paradigm that researchers working on vision research should focus on.
In his talk, Dr. Shum will also give a quick overview of Microsoft Research Asia, “the world’s hottest computer lab” (MIT Technology Review, June 2004), and outline its activities in basic research, technology transfer, product incubation, technology licensing, university relations and internship program.
This event is free of charge.
For more information, please visit:
Should you have any enquiries, please feel free to contact Faculty of Engineering by email at firstname.lastname@example.org or by phone at 28592803 or by fax at 25469142.
ALL ARE WELCOME.