ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extraction

Page view(s)
76
Checked on Nov 29, 2024
ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extraction
Title:
ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extraction
Journal Title:
IEEE Transactions on Multimedia
Publication Date:
03 March 2022
Citation:
Zhao, X., Wu, X., Miao, J., Chen, W., Chen, P. C. Y., & Li, Z. (2022). ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extraction. IEEE Transactions on Multimedia, 1–1. https://doi.org/10.1109/tmm.2022.3155927
Abstract:
Existing methods detect the keypoints in a nondifferentiable way, therefore they can not directly optimize the position of keypoints through back-propagation. To address this issue, we present a partially differentiable keypoint detection module, which outputs accurate sub-pixel keypoints. The reprojection loss is then proposed to directly optimize these sub-pixel keypoints, and the dispersity peak loss is presented for accurate keypoints regularization. We also extract the descriptors in a subpixel way, and they are trained with the stable neural reprojection error loss. Moreover, a lightweight network is designed for keypoint detection and descriptor extraction, which can run at 95 frames per second for 640x480 images on a commercial GPU. On homography estimation, camera pose estimation, and visual (re-)localization tasks, the proposed method achieves equivalent performance with the state-of-the-art approaches, while greatly reduces the inference time.
License type:
Publisher Copyright
Funding Info:
This work was supported by the National Nature Science Foundation of China under Grant No. 61620106012
Description:
© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
ISSN:
1520-9210
1941-0077
Files uploaded:

File Size Format Action
alike-v0204.pdf 7.07 MB PDF Open