LPRnet: A self-supervised registration network for LiDAR and photogrammetric point clouds
This work addresses the integration of complementary remote sensing data for applications like mapping and surveying, but it is incremental as it builds on existing self-supervised and transformer-based methods.
The paper tackles the problem of registering heterogeneous LiDAR and photogrammetric point clouds, which have significant discrepancies and lack ground truth, by proposing a self-supervised registration network based on a masked autoencoder; experiments on real-world datasets validate its effectiveness in solving this registration challenge.
LiDAR and photogrammetry are active and passive remote sensing techniques for point cloud acquisition, respectively, offering complementary advantages and heterogeneous. Due to the fundamental differences in sensing mechanisms, spatial distributions and coordinate systems, their point clouds exhibit significant discrepancies in density, precision, noise, and overlap. Coupled with the lack of ground truth for large-scale scenes, integrating the heterogeneous point clouds is a highly challenging task. This paper proposes a self-supervised registration network based on a masked autoencoder, focusing on heterogeneous LiDAR and photogrammetric point clouds. At its core, the method introduces a multi-scale masked training strategy to extract robust features from heterogeneous point clouds under self-supervision. To further enhance registration performance, a rotation-translation embedding module is designed to effectively capture the key features essential for accurate rigid transformations. Building upon the robust representations, a transformer-based architecture seamlessly integrates local and global features, fostering precise alignment across diverse point cloud datasets. The proposed method demonstrates strong feature extraction capabilities for both LiDAR and photogrammetric point clouds, addressing the challenges of acquiring ground truth at the scene level. Experiments conducted on two real-world datasets validate the effectiveness of the proposed method in solving heterogeneous point cloud registration problems.