metadata
license: cc-by-nc-nd-4.0
Dataset for monocular visual relocalization on a COLMAP 3D reconstruction model.
This dataset was collected at EZOHUB Tokyo with the cooperation of SATUDORA HOLDINGS CO.,LTD.
Dataset structure
train: Images used to construct the COLMAP model.model: The COLMAP model files.test: Test images to run relocalization.intrinsic.xml: Intrinsic camera parameters.