Geo-Supervised Visual Depth Prediction

07/30/2018
by   Xiaohan Fei, et al.
8

We propose using global orientation from inertial measurements, and the bias it induces on the shape of objects populating the scene, to inform visual 3D reconstruction. We test the effect of using the resulting prior in depth prediction from a single image, where the normal vectors to surfaces of objects of certain classes tend to align with gravity or be orthogonal to it. Adding such a prior to baseline methods for monocular depth prediction yields improvements beyond the state-of-the-art and illustrates the power of gravity as a supervisory signal.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset