Pictorial Representation of the Basic Stereo Matching Process in Computer Vision
figureposted on 11.08.2020 by James Cooper
Figures are generally photos, graphs and static images that would be represented in traditional pdf publications.
Diagram of the basic process of Stereo Matching. In this instance, for each pixel in the left image, a horizontal range of the pixels in the right image are searched, to find the one on the right that matches most closely to the one on the left. This has the effect that the disparity map is from the perspective of the left camera. The red dots represent the compared pixels. The brown dashed arrow and vertical bars represent the range of pixels in the right image to compute matching scores against. The length of the range is the lesser of the number of pixels before reaching the left border of the image, or the `maximum disparity', which is a parameter set by the user and represented here by Λ.
Both formats are the same image, they are both included for your convenience, and because figshare apparently doesn't render .eps files in-browser but .eps seems to be the required vector graphics format for many publications.