¹ Hefei University of Technology, Hefei , P.R. China
² University of Science and Technology of China, Hefei, P.R. China

---

Abstract—In this paper, we address a challenging problem of depth estimation taken by an uncalibrated conventional camera based on defocus images. In order to be a solution to the traditional method may not apply to the natural scene, we make some effective improvement on it. And we also take guided filtering to repair our depth map. Experimental results on a variety of images show that our method can acquire visual pleasing results. Moreover, our method can also extract the depth map of the scene with fairly good extent of accuracy.

Index Terms—depth estimation, guided filtering, depth from defocus, uncalibrated, defocus images

[PDF]

Cite: Hao Wang, Fengyun Cao, Shuai Fang, Yang Cao, and Chunlong Fang, " Effective Improvement for Depth Estimated Based on Defocus Images," Journal of Computers vol. 8, no. 4, pp. 888-895, 2013.