Visual depth perception based on optical blur

The depth perception we present in this paper is based on monocular computer vision. The method exploits the physical effect that the imaging properties of an optical system depend upon the acquisition parameters and the object distance. Our approach is working along the edges. The basic principle is to compare the blur in two defocused images of the same scene taken with different apertures (depth from defocus). To increase speed and precision our algorithm is working only at the exact position of ramp edges, which are determined by a biological model of the visual cortex.