Tuesday, 20.9.2016, 12:00
Computer science and optics are usually studied separately -- separate people, in separate departments, meet at separate conferences. This is changing. The exciting promise of technologies like virtual reality and self-driving cars demand solutions that draw from the best aspects of computer vision, computer graphics, and optics. Previously, it has proved difficult to bridge these communities. For instance, the laboratory setups in optics are often designed to image millimeter-size scenes in a vibration-free darkroom.
In this talk, I show approaches that bridge the divide between optics and computer science. This involves the use of slightly more complicated optics than one would ordinarily use in vision and graphics. For example, time of flight 3D cameras offer more information than ordinary 2D cameras, but remain a commodity device. The extra bit of temporal information, when used with joint computational processing, can advance new forms of imaging, like seeing around corners or imaging through scattering media. In another example, a commodity polarizing filter is used to analyze different rotations of light waves to create a 3D depth camera that is far more accurate than the Microsoft Kinect. We discuss the broader impact of this design paradigm on the future of depth sensors, interferometers, computational photography, medical imaging, HCI and many other applications
Achuta Kadambi is a PhD student at the Massachusetts Institute of Technology, working at the intersection of computer science and optics. A key aim is to overcome traditional challenges in computer vision by leveraging joint design of optical capture and computational processing. Kadambi has taught courses on computational photography and light transport at SIGGRAPH and ICCV and co-lectured an MIT class titled "mathematical methods in imaging". During his PhD he has interned at Microsoft Research in Seattle WA and Mitsubishi Electric (MERL) in Cambridge MA. He has won several awards, most recently, the 2016 Lemelson-MIT student prize. His website is here: http://web.media.mit.edu/~achoo/