期刊
REMOTE SENSING
卷 14, 期 11, 页码 -出版社
MDPI
DOI: 10.3390/rs14112559
关键词
virtual points; partial-to-partial; transformer; GAN
With the development of scanning technology, point cloud registration plays a significant role in various fields. Traditional algorithms like ICP search for corresponding points based on the closest point, while recent deep learning-based algorithms utilize deep features for point calculation. However, the partiality of point clouds poses challenges for partial-to-partial registration. To address this, we propose VPRNet, which combines a virtual point generation network with a registration network to overcome the limitations of partiality. The experiments demonstrate the superior performance of our proposed algorithm compared to existing methods.
With the development of high-precision and high-frame-rate scanning technology, we can quickly obtain scan data of various large-scale scenes. As a manifestation of information fusion, point cloud registration is of great significance in various fields, such as medical imaging, autonomous driving, and 3D reconstruction. The Iterative Closest Point (ICP) algorithm, as the most classic algorithm, leverages the closest point to search corresponding points, which is the pioneer of correspondences-based approaches. Recently, some deep learning-based algorithms witnessed extracting deep features to compress point cloud information, then calculate corresponding points, and finally output the optimal rigid transformation like Deep Closest Point (DCP) and DeepVCP. However, the partiality of point clouds hinders the acquisition of enough corresponding points when dealing with the partial-to-partial registration problem. To this end, we propose Virtual Points Registration Network (VPRNet) for this intractable problem. We first design a self-supervised virtual point generation network (VPGnet), which utilizes the attention mechanism of Transformer and Self-Attention to fuse the geometric information of two partial point clouds, combined with the Generative Adversarial Network (GAN) structure to produce missing points. Subsequently, the following registration network structure is spliced to the end of VPGnet, thus estimating rich corresponding points. Unlike the existing methods, our network tries to eliminate the side effects of incompleteness on registration. Thus, our method expresses resilience to the initial rotation and sparsity. Various experiments indicate that our proposed algorithm shows advanced performance compared to recent deep learning-based and classical methods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据