
Abstract

This paper presents a fast algorithm for deriving the defocus map from a single image. Existing methods of defocus map estimation often include a pixellevel propagation step to spread the measured sparse defocus cues over the whole image. Since the pixellevel propagation step is timeconsuming, we develop an effective method to obtain the wholeimage defocus blur using oversegmentation and
transductive inference. Oversegmentation produces the superpixels and hence greatly reduces the computation costs for subsequent procedures. Transductive inference provides a way to calculate the similarity between superpixels, and thus helps to infer the defocus blur of each superpixel from all other superpixels. The experimental results show that our method is efficient and able to estimate a plausible superpixellevel defocus map from a given single image.
Last updated: 14 October 2016