Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters
TL
TL;DR
A novel approach for keypoint detection task that combines handcrafted and learned CNN filters within a shallow multi-scale architecture that outperforms state-of-the-art detectors in terms of repeatability, matching performance and complexity is introduced.
Abstract
We introduce a novel approach for keypoint detection task that combines handcrafted and learned CNN filters within a shallow multi-scale architecture. Handcrafted filters provide anchor structures for learned filters, which localize, score and rank repeatable features. Scale-space representation is used within the network to extract keypoints at different levels. We design a loss function to detect robust features that exist across a range of scales and to maximize the repeatability score. Our Key.Net model is trained on data synthetically created from ImageNet and evaluated on HPatches benchmark. Results show that our approach outperforms state-of-the-art detectors in terms of repeatability, matching performance and complexity.
References45 items
1
Large scale evaluation of local image feature detectors on homography datasets
2
Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning
3
LF-Net: Learning Local Features from Images
4
End-to-End Learning of Keypoint Detector and Descriptor for Pose Invariant 3D Matching
5