Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters
arXiv:1904.00889
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.
References in corpus (6)
- Faster and better: a machine learning approach to corner detection
- TILDE: A Temporally Invariant Learned DEtector
- LF-Net: Learning Local Features from Images
- Working hard to know your neighbor's margins: Local descriptor learning loss
- Learning to Compare Image Patches via Convolutional Neural Networks
- Toward Geometric Deep SLAM
Cited by in corpus (7)
- R2D2: Repeatable and Reliable Detector and Descriptor
- A Survey on Deep Learning for Localization and Mapping: Towards the Age of Spatial Machine Intelligence
- Unifying Deep Local and Global Features for Image Search
- UR2KiD: Unifying Retrieval, Keypoint Detection, and Keypoint Description without Local Correspondence Supervision
- CAE-LO: LiDAR Odometry Leveraging Fully Unsupervised Convolutional Auto-Encoder for Interest Point Detection and Feature Description
- Under the Radar: Learning to Predict Robust Keypoints for Odometry Estimation and Metric Localisation in Radar
- D2D: Keypoint Extraction with Describe to Detect Approach