《Single Image Focus Editing》.pdf

Single Image Focus Editing Wei Zhang and Wai-Kuen Cham Department of Electronic Engineering The Chinese University of Hong Kong, Shatin, N.T., Hong Kong {zhangwei,wkcham}@.hk Abstract In this paper, we present a postprocessing approach to tackle the single image focus editing problem. In detail, the proposed method can accomplish the tasks of focus map (a) (b) (c) estimation, image refocusing and defocusing. Given an im- age with a mixture of focused and defocused objects, we first detect the edges and then estimate the focus map based on the edge blurriness which is depicted explicitly with a well-parameterized model. The image refocusing problem (d) (e) (f) is addressed in an elaborate blind deconvolution frame- work, where the image prior is modeled well by using both Figure 1. (a) The input photo captured with narrow aperture and global and local constraints. Especially, we correct the de- focused on the foreground object; (b) The synthesized photo with focused blurry edges to sharp ones with the aid of the para- shallower DOF; (c) The synthesized all-focused photo; (d) The metric edge model and then render this cue as a novel local synthesized photo focused on the background; (e) The detected prior to ensure the sharpness of the refocused image. Ex- focus mask. (white: defocused regions; black: focused regions; gray: focus boundaries)