Image Deblurring with Coded Exposure (Flutter Shutter Camera)
A new camera, flutter shutter camera is proposed to aid motion deblurring due to fast moving objects (cars etc) or cameras (aerial imaging). The modification amounts to simply opening and closing the shutter according to a pseudo-random code during the exposure time. The flutter shutter camera is also used for super-resolution of moving objects by utilizing the motion blur information.
Background & Objective: Objective: Image deblurring is an ill-posed problem and traditional methods lead to noise and ringing artifacts in the deblurred image. The proposed camera can make deblurring a well-posed problem.
Technical Discussion: A traditional camera loses high spatial frequencies if there is a relative motion between the camera and the object during the exposure time. By opening and closing the shutter using carefully chosen binary codes, high spatial frequencies can be maintained. This makes motion deblurring a well-posed problem.
Outside Collaborations: Prof. Jack Tumblin, Dept of Computer Science, Northwestern University.
Future Direction: 1) Investigate image priors for reducing noise in reconstruction; 2) Investigate continuous valued code and better codes by accounting for CCD photon noise; and 3) Utilize Flutter shutter videos for automatic motion estimation and deblurring.
Contacts:
Ramesh Raskar
Amit Agrawal
Publications:
Agrawal, A.; Raskar, R., "Resolving Objects at Higher Resolution from a Single Motion-blurred Image", IEEE Computer Society Conference on Computer Vision & Pattern Recognition (CVPR), ISBN: 1-4244-1180-7, pp. 1-8, June 2007 (IEEE Xplore, TR2007-036)
Raskar, R.; Agrawal, A.; Tumblin, J., "Coded Exposure Photography: Motion Deblurring Using Fluttered Shutter", ACM Transactions on Graphics (TOG), ISSN: 0730-0301, Vol. 25, Issue 3, pp. 795-804, July 2006 (TR2006-124)
Technology Areas:
Imaging
Computer Vision
Modification Date: September 12, 2007
