SfM-TTR: Using Structure from Motion for Test-Time Refinement of Single-View Depth Networks.

Sergio Izquierdo, Javier Civera

Browse the full CVPR paper archive.