MD-Net: Multi-Detector for Local Feature Extraction

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung


Establishing a sparse set of keypoint correspondences between images is a fundamental task in many computer vision pipelines. Often, this translates into a computationally expensive nearest neighbor search, where every keypoint descriptor at one image must be compared with all the descriptors at the others. In order to lower the computational cost of the matching phase, we propose a deep feature extraction network capable of detecting a predefined number of complementary sets of keypoints at each image. Since only the descriptors within the same set need to be compared across the different images, the matching phase computational complexity decreases with the number of sets. We train our network to predict the keypoints and compute the corresponding descriptors jointly. In particular, in order to learn complementary sets of keypoints, we introduce a novel unsupervised loss which penalizes intersections among the different sets. Additionally, we propose a novel descriptor-based weighting scheme meant to penalize the detection of keypoints with non-discriminative descriptors. With extensive experiments we show that our feature extraction network, trained only on synthetically warped images and in a fully unsupervised manner, achieves competitive results on 3D reconstruction and re-localization tasks at a reduced matching complexity.

Titel2022 26th International Conference on Pattern Recognition, ICPR 2022
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers
ISBN (elektronisch)9781665490627
PublikationsstatusVeröffentlicht - 2022
Veranstaltung26th International Conference on Pattern Recognition: ICPR 2022 - Montreal, Kanada
Dauer: 21 Aug. 202225 Aug. 2022


Konferenz26th International Conference on Pattern Recognition
KurztitelICPR 2022

ASJC Scopus subject areas

  • Maschinelles Sehen und Mustererkennung


Untersuchen Sie die Forschungsthemen von „MD-Net: Multi-Detector for Local Feature Extraction“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren