Fig. 5
From: Understanding and simulating border ownership centered segmentation
Binocular Vision Examples (image from Virtual Kitti 2 dataset, Cabon et al., 2020)5. The first two images are the stereo left and right images, respectively; the 3rd image is a stereo anaglyph combining the two; the 4th image is a 2-channel border ownership coding map when viewed from the left camera; the 5th image is the ‘Near’ (Epipolar) disparity map where the greater depth appear lighter in red; the 6th image is the ‘Far’ (logarithmic) disparity map, also using red for depth coding, preserving depth order across distances.
