2018
ActiveStereoNet: End-to-End Self-Supervised Learning for Active Stereo Systems
ECCV 2018poster
In this paper we present ActiveStereoNet, the first deep learning solution for active stereo systems. Due to the lack of ground truth, our method is fully self-supervised, yet it produces precise depth with a subpixel precision of 1/30th of a pixel; it does not suffer from the common over-smoothing…