Fetching the paper…

Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video · Around