Submission details

Method nameDenseVLAD & D2-Net (top-20)
Datasetaachen
Submission dateApril 11, 2019, 11:13 a.m.
Last modifiedMay 29, 2019, 7:48 a.m.
Publication urlDusmanu et al., D2-Net: A Trainable CNN for Joint Description and Detection of Local Features, CVPR 2019
Code urlhttps://github.com/mihaidusmanu/d2-net
Other infoUses DenseVLAD to find the top-20 database images for each query, matches D2-Net features between the query and database images, then uses COLMAP's image_registrator to estimate the camera poses.

Detailed results for all conditions

Methoddaynight
DenseVLAD & D2-Net (top-20)80.1 / 88.0 / 93.439.8 / 55.1 / 74.5