Abstract:

Sharing what we see with computers will open the door for novel services that are both personalized and crowd-powered. We envision a wearable camera system captures vision data continuously in our daily life and feeds it to third-party applications. In this talk, we show that today’s system software and imaging hardware, highly optimized for photography, are fundamentally ill-suited for this. We present system software support for efficient concurrency and show power-saving mechanisms available in existing imaging hardware can go a long way yet are not enough. In the end, we argue that a fundamental rethinking of the imaging pipeline is necessary toward realizing continuous mobile vision.

Bio:

Lin Zhong received his B.S and M.S. from Tsinghua University and Ph.D. from Princeton University. He has been with Rice University since September 2005 where he is currently an associate professor. He was a visiting researcher with Microsoft Research for the summer of 2011 and March to December 2012, where his work on continuous mobile vision started. At Rice, he leads the Efficient Computing Group to make computing, communication, and interfacing more efficient and effective. He and his students received the best paper awards from ACM MobileHCI 2007, IEEE PerCom 2009, and ACM MobiSys 2011, 2013 and 2014, and ACM ASPLOS 2014.