TW-SMNet: Deep Multitask Learning of Tele-Wide Stereo Matching

November 2019

tl;dr: Depth estimation by fusing the images from two cameras with different focal length.

Overall impression

The paper is among the first to fuse stereo pairs with different focal length. For the intersection FoV (tele FoV), we can resize the wide FoV image to match that of the tele FoV, reducing the problem to normal stereo pair matching algorithm.

Key ideas

Technical details