Learning to Produce Semi-Dense Correspondences for Visual Localization.

Khang Truong Giang, Soohwan Song, Sungho Jo

Browse the full CVPR paper archive.