Submission details
Method name | DenseVLAD & D2-Net (top-20) |
Dataset | aachen |
Submission date | April 11, 2019, 11:13 a.m. |
Last modified | Oct. 31, 2019, 12:58 a.m. |
Publication url | http://openaccess.thecvf.com/content_CVPR_2019/papers/Dusmanu_D2-Net_A_Trainable_CNN_for_Joint_Description_and_Detection_of_CVPR_2019_paper.pdf |
Code url | https://github.com/mihaidusmanu/d2-net |
Other info | Uses 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
Method | day | night |
---|---|---|
DenseVLAD & D2-Net (top-20) | 80.1 / 88.0 / 93.4 | 66.3 / 75.5 / 84.7 |