A Novel Object-Level Building-Matching Method across 2D Images and 3D Point Clouds Based on the Signed Distance Descriptor (SDD)
-
Published:2023-06-07
Issue:12
Volume:15
Page:2974
-
ISSN:2072-4292
-
Container-title:Remote Sensing
-
language:en
-
Short-container-title:Remote Sensing
Author:
Zhao Chunhui12, Wang Wenxuan12, Yan Yiming12ORCID, Su Nan12ORCID, Feng Shou12ORCID, Hou Wei3, Xia Qingyu12
Affiliation:
1. Key Laboratory of Advanced Marine Communication and Information Technology, Ministry of Industry and Information, Harbin 150009, China 2. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150009, China 3. Harbin Aerospace Star Data System Science and Technology Co., Ltd., Harbin 150028, China
Abstract
In this work, a novel object-level building-matching method using cross-dimensional data, including 2D images and 3D point clouds, is proposed. The core of this method is a newly proposed plug-and-play Joint Descriptor Extraction Module (JDEM) that is used to extract descriptors containing buildings’ three-dimensional shape information from object-level remote sensing data of different dimensions for matching. The descriptor is named Signed Distance Descriptor (SDD). Due to differences in the inherent properties of different dimensional data, it is challenging to match buildings’ 2D images and 3D point clouds on the object level. In addition, features extracted from the same building in images taken at different angles are usually not exactly identical, which will also affect the accuracy of cross-dimensional matching. Therefore, the question of how to extract accurate, effective, and robust joint descriptors is key to cross-dimensional matching. Our JDEM maps different dimensions of data to the same 3D descriptor SDD space through the 3D geometric invariance of buildings. In addition, Multi-View Adaptive Loss (MAL), proposed in this paper, aims to improve the adaptability of the image encoder module to images with different angles and enhance the robustness of the joint descriptors. Moreover, a cross-dimensional object-level data set was created to verify the effectiveness of our method. The data set contains multi-angle optical images, point clouds, and the corresponding 3D models of more than 400 buildings. A large number of experimental results show that our object-level cross-dimensional matching method achieves state-of-the-art outcomes.
Funder
National Natural Science Foundation of China Heilongjiang Outstanding Youth Foundation Heilongjiang Postdoctoral Foundation Fundamental Research Funds for the Central Universities Grant High-Resolution Earth Observation Major Project
Subject
General Earth and Planetary Sciences
Reference51 articles.
1. Liu, L., Li, H., and Dai, Y. (2017, January 22–29). Efficient global 2D-3D matching for camera localization in a large-scale 3D map. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy. 2. Efficient & effective prioritized matching for large-scale image-based localization;Sattler;IEEE Trans. Pattern Anal. Mach. Intell.,2016 3. Song, Y., Chen, X., Wang, X., Zhang, Y., and Li, J. (2017, January 21–26). Are large-scale 3-D models really necessary for accurate visual localization?. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA. 4. Kundu, J.N., Rahul, M.V., Ganeshan, A., and Babu, R.V. (2018, January 8–14). Object pose estimation from monocular image using multi-view keypoint correspondence. Proceedings of the European Conference on Computer Vision, Munich, Germany. 5. MonoSLAM: Real-time single camera SLAM;Davison;IEEE Trans. Pattern Anal. Mach. Intell.,2007
|
|