Kpsnet: Keypoint Detection and Feature Extraction for Point Cloud Registration

Du, A; Huang, X; Zhang, J; Yao, L; Wu, Q

Kpsnet: Keypoint Detection and Feature Extraction for Point Cloud Registration

Du, A

Huang, X

Zhang, J

Yao, L Wu, Q

Permalink

Publication Type:: Conference Proceeding
Citation:: Proceedings - International Conference on Image Processing, ICIP, 2019, 2019-September pp. 2576 - 2580
Issue Date:: 2019-09-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

The embargo period expires on 1 Sep 2021

Adobe PDF

Download Accepted Manuscript VersionAdobe PDF (610.95 kB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Du, A https://orcid.org/0000-0002-9634-3125	en_US
dc.contributor.author	Huang, X https://orcid.org/0000-0002-3579-538X	en_US
dc.contributor.author	Zhang, J https://orcid.org/0000-0002-7240-3541	en_US
dc.contributor.author	Yao, L	en_US
dc.contributor.author	Wu, Q https://orcid.org/0000-0001-5641-2483	en_US
dc.date.issued	2019-09-01	en_US
dc.identifier.citation	Proceedings - International Conference on Image Processing, ICIP, 2019, 2019-September pp. 2576 - 2580	en_US
dc.identifier.isbn	9781538662496	en_US
dc.identifier.issn	1522-4880	en_US
dc.identifier.uri	http://hdl.handle.net/10453/137262
dc.description.abstract	© 2019 IEEE. This paper presents the KPSNet, a KeyPoint Siamese Network to simultaneously learn task-desirable keypoint detector and feature extractor. The keypoint detector is optimized to predict a score vector, which signifies the probability of each candidate being a keypoint. The feature extractor is optimized to learn robust features of keypoints by exploiting the correspondence between the keypoints generated from two inputs, respectively. For training, the KPSNet does not require to manually annotate keypoints and local patches pairwise. Instead, we design an alignment module to establish the correspondence between the two inputs and generate positive and negative samples on-the-fly. Therefore, our method can be easily extended to new scenes. We test the proposed method on the open-source benchmark and experiments show the validity of our method.	en_US
dc.relation.ispartof	Proceedings - International Conference on Image Processing, ICIP	en_US
dc.relation.isbasedon	10.1109/ICIP.2019.8803365	en_US
dc.rights	info:eu-repo/semantics/embargoedAccess
dc.title	Kpsnet: Keypoint Detection and Feature Extraction for Point Cloud Registration	en_US
dc.type	Conference Proceeding
utslib.citation.volume	2019-September	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
pubs.organisational-group	/University of Technology Sydney/Strength - INEXT - Innovation in IT Services and Applications
pubs.organisational-group	/University of Technology Sydney/Students
utslib.copyright.status	open_access	*
utslib.copyright.embargo	2021-09-01T00:00:00+1000
pubs.publication-status	Published	en_US
pubs.volume	2019-September	en_US

Abstract:

© 2019 IEEE. This paper presents the KPSNet, a KeyPoint Siamese Network to simultaneously learn task-desirable keypoint detector and feature extractor. The keypoint detector is optimized to predict a score vector, which signifies the probability of each candidate being a keypoint. The feature extractor is optimized to learn robust features of keypoints by exploiting the correspondence between the keypoints generated from two inputs, respectively. For training, the KPSNet does not require to manually annotate keypoints and local patches pairwise. Instead, we design an alignment module to establish the correspondence between the two inputs and generate positive and negative samples on-the-fly. Therefore, our method can be easily extended to new scenes. We test the proposed method on the open-source benchmark and experiments show the validity of our method.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/137262