Deep learning has made significant impacts on multi-view stereo systems....
We study the task of semantic mapping - specifically, an embodied agent ...
We develop new representations and algorithms for three-dimensional (3D)...
Passive visual systems typically fail to recognize objects in the amodal...
We tackle the problem of automatically reconstructing a complete 3D mode...
To date, top-performing optical flow estimation methods only take pairs ...
Given two consecutive frames from a pair of stereo cameras, 3D scene flo...