Submission details

Method nameDenseVLAD & D2-Net (top-20)
Datasetaachen
Submission dateApril 11, 2019, 11:13 a.m.
Last modifiedOct. 31, 2019, 12:58 a.m.
Publication urlhttp://openaccess.thecvf.com/content_CVPR_2019/papers/Dusmanu_D2-Net_A_Trainable_CNN_for_Joint_Description_and_Detection_of_CVPR_2019_paper.pdf
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 / 93.466.3 / 75.5 / 84.7