“Computational Cameras and Displays” by O’Toole and Wetzstein

  • ©Matthew O’Toole and Gordon Wetzstein

Conference:


Type:


Entry Number: 09

Title:

    Computational Cameras and Displays

Course Organizer(s):



Presenter(s)/Author(s):



Abstract:


    Abstract
    Recent advancements in both computational photography and displays  have given rise to a new generation of computational devices. These computational cameras and displays provide a visual experience that go beyond the capabilities of traditional systems, by adding computational power  to optics, lights, and sensors. This technology is breaking new ground  in the consumer market with lightfield cameras that redefines our under- standing of pictures (Lytro), displays for visualizing 4D content without  special eyewear (Nintendo 3DS), motion sensing devices that use light coded in space or time to detect motion and position (Kinect, Leap Motion), and a movement towards ubiquitous computing through wearable cameras and displays (Google Glass). 

    This course provides an introduction to the state-of-the-art in computational cameras and displays. The objective of this course is to give the  audience a broad overview of key concepts and works that could benefit their own research endeavors in computer graphics and vision. In contrast to past courses and papers that focus on either imaging or displays, this course combines both subject matters to highlight the duality of the principles behind the technology, and the combination of such cameras and  displays to infer properties of real, unknown, and complex scenes. 

    We begin by introducing the three key components in any computational camera or display: lights, sensors, and optics. Our talk then focuses  on computational imaging, where the objective is to capture new forms of  visual information by adding computation to cameras. The course transitions into a detailed overview of displays, focusing on autostereoscopic  displays and lightfield projectors. This leads into a discussion on computational light transport, an area that combines computational cameras and  lights to analyze the light transport of real-world scenes, with a broad range of applications that includes image-based relighting, capturing geometry, and visualizing light transport phenomena.  


Overview Page: