Neural-Guided RANSAC for Estimating Epipolar Geometry [Data]

DOI

Pre-computed sparse feature correspondences for pairs of images (outdoor and indoor) to reproduce the experiments described in our paper, particularly to train and evaluate NG-RANSAC.

For more information, also see the code documentation: https://github.com/vislearn/ngransac

Identifier
DOI https://doi.org/10.11588/data/PCGYET
Metadata Access https://heidata.uni-heidelberg.de/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.11588/data/PCGYET
Provenance
Creator Brachmann, Eric
Publisher heiDATA
Contributor Brachmann, Eric; heiDATA: Heidelberg Research Data Repository
Publication Year 2020
Rights info:eu-repo/semantics/openAccess
OpenAccess true
Contact Brachmann, Eric (Technische Universität Dresden)
Representation
Resource Type Dataset
Format application/gzip
Size 3981726462
Version 1.0
Discipline Other
Spatial Coverage Hannover