D2-Net: A Trainable CNN for Joint Detection and Description of Local Features - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

D2-Net: A Trainable CNN for Joint Detection and Description of Local Features

Résumé

In this work we address the problem of finding reliable pixel-level correspondences under difficult imaging conditions. We propose an approach where a single convolutional neural network plays a dual role: It is simultaneously a dense feature descriptor and a feature detector. By postponing the detection to a later stage, the obtained keypoints are more stable than their traditional counterparts based on early detection of low-level structures. We show that this model can be trained using pixel correspondences extracted from readily available large-scale SfM reconstructions, without any further annotations. The proposed method obtains state-of-the-art performance on both the difficult Aachen Day-Night localization dataset and the InLoc indoor localization benchmark, as well as competitive performance on other benchmarks for image matching and 3D reconstruction.

Dates et versions

hal-02438461 , version 1 (14-01-2020)

Identifiants

Citer

Mihai Dusmanu, Ignacio Rocco, Tomas Pajdla, Marc Pollefeys, Josef Sivic, et al.. D2-Net: A Trainable CNN for Joint Detection and Description of Local Features. CVPR 2019 - IEEE Conference on Computer Vision and Pattern Recognition, Jun 2019, Long Beach, United States. ⟨hal-02438461⟩
314 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More