“Motion regularization for matting motion blurred objects” by , Tai and Brown

  • ©Lin Hai TIng, Yu-Wing Tai, and Michael S. Brown

Conference:


Type:


Title:

    Motion regularization for matting motion blurred objects

Presenter(s)/Author(s):



Abstract:


    We address the problem of matting motion blurred objects from a single image. Existing single-image matting methods are designed to extract static objects that have fractional pixel occupancy. This arises because the real scene object has a finer resolution than the discrete image pixel and therefore only occupies a portion of the pixel. For a motion blurred object, however, fractional pixel occupancy is attributed almost entirely to the object’s motion over the exposure time. While conventional matting techniques can be used to matte motion blurred object, they are not formulated in a manner that considers the object’s local motion. Not surprisingly, these existing techniques often produce less than satisfactory results when used to matte motion blurred objects, especially when not on solid colored background.

References:


    1. Levin, A., Lischinski, D., and Weiss, Y. 2006. A closed form solution to natural image matting. In CVPR’06.
    2. Levin, A. 2006. Blind motion deblurring using image statistics. In NIPS’06.
    3. Liu, R., Li, Z., and Jia, J. 2008. Image partial blur detection and classification. In CVPR’08.
    4. Wang, J., and Cohen, M. 2007. Optimized color sampling for robust matting. In CVPR’07.
    5. Wang, J., Agrawala, M., and Cohen, M. 2007. Soft scissors: An interactive tool for realtime high quality matting.


ACM Digital Library Publication:



Overview Page: