Over a decade has passed since the inception of computational photography—the technology that combines computing, digital sensors, modern optics and actuators to escape the limitations of traditional cameras—but has the technology delivered on its promise? In the last ten years the field has delivered light-field cameras, multi-spectral cameras, programmable flash and post-capture control to consumers, culminating in new visual experiences. Like many products, however, the evolution has followed the Kano model, which stipulates that over time, innovative technologies are taken for granted and become a basic need. Though there have been many successes like HDR photography, panorama stitching and 3D imaging, the dreams have been much more ambitious than the successes. As computational photography progresses, however, I predict that we move to research in the area of hacking physics—ultrafast cameras, time-of-flight technology, light field sensors, and terahertz imaging—or hacking experiences—virtual reality displays (Oculus Rift, Microsoft HoloLens), drone imaging, end-user ultrasound / X-ray and health applications. These innovations will put the lost decade of computational photography behind, and usher in a glorious one of superhuman vision.

Presenter Bio

Ramesh Raskar, MIT Media Lab