26 June 2021 PC-SuperPoint: interest point detection and descriptor extraction using pyramid convolution and circle loss
Yu-Jie Xiong, Shuo Ma, Yongbin Gao, Zhijun Fang
Author Affiliations +
Abstract

Nowadays, deep learning is widely used to detect interest points and extract the corresponding descriptors and achieved suitable results for many applications of computer vision, such as image matching, three-dimensional reconstruction, simultaneous localization, and mapping. We propose an approach for interest point detection and descriptor extraction using pyramid convolution and circle loss, which is named as PC-SuperPoint. We utilize pyramid convolutions in the backbone network, which includes convolution kernels of different scales for multiscale feature extraction. The following well-designed networks are able to capture the local and global information from the obtained backbone feature maps. In addition, circle loss, which enhances weight attributes for each pair of descriptors, is also applied to improve the convergence speed in the training phase. Experiments on the HPatches dataset and KITTI dataset achieve promising results, which reveal the effectiveness of the proposed method.

© 2021 SPIE and IS&T 1017-9909/2021/$28.00 © 2021 SPIE and IS&T
Yu-Jie Xiong, Shuo Ma, Yongbin Gao, and Zhijun Fang "PC-SuperPoint: interest point detection and descriptor extraction using pyramid convolution and circle loss," Journal of Electronic Imaging 30(3), 033024 (26 June 2021). https://doi.org/10.1117/1.JEI.30.3.033024
Received: 28 October 2020; Accepted: 8 June 2021; Published: 26 June 2021
Lens.org Logo
CITATIONS
Cited by 3 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Convolution

Feature extraction

Visualization

Sensors

Error analysis

Network architectures

Data modeling

Back to Top