Deep Global Features for Point Cloud Alignment

Abstract

Point cloud registration is a key problem in computer vision applications and involves finding a rigid transform from a point cloud into another such that they align together. The iterative closest point (ICP) method is a simple and effective solution that converges to a local optimum. However, despite the fact that point cloud registration or alignment is addressed in learning-based methods, such as PointNetLK, they do not offer good generalizability for point clouds. In this stud, we proposed a learning-based approach that addressed existing problems, such as finding local optima for ICP and achieving minimum generalizability. The proposed model consisted of three main parts: an encoding network, an auxiliary module that weighed the contribution of each input point cloud, and feature alignment to achieve the final transform. The proposed architecture offered greater generalization among the categories. Experiments were performed on ModelNet40 with different configurations and the results indicated that the proposed approach significantly outperformed the state-of-the-art point cloud alignment methods.

Cascade Convolution Neural Network for Point Set Generation

“…In this work, we build a cascade network to generate a more flexible and dense point cloud from single-view image. Compared with previous research [5], we generate more uniform and dense point cloud. However, there are still some details lost, especially in the edge of point cloud.…”

“…To project 3D points, we applied perspective projection in this section. In detail, we defined a point 𝑑𝑑 𝑥𝑥,𝑦𝑦,𝑧𝑧 as the transformation of point 𝑝𝑝 𝑥𝑥,𝑦𝑦,𝑧𝑧 with the 3D world coordinate to the camera coordinates system using Equation (5).…”

“…Although all three types of representation can effectively represent and visualize 3D objects, point cloud provides more advantages in industrial applications that require special geometrical information, and offers a higher degree of freedom with less memory [3]. More specifically, voxel-based representation takes an enormous amount of memory with low computational efficiency, especially at high resolution [4], and mesh-based representation indicates poor flexibility and scalability since it is commonly derived from point clouds [5]. Furthermore, point cloud is a standard 3D acquisition format that is especially suitable for geometric operations and transformation.…”

“…The main idea is that multiple view geometry [7] recovers the 3D geometric structure information of the object from images of the static object from different viewpoints. The common approaches include structure from motion [8], simultaneous localization and mapping [9], scene reconstruction [10], and point cloud alignment [5]. Compared with multiple images 3D reconstruction, it is more difficult to predict 3D geometric information from a single-view image.…”

Robust Coarse-to-Fine Registration Scheme for Mobile Laser Scanner Point Clouds Using Multiscale Eigenvalue Statistic-Based Descriptor

“…Compared with traditional descriptors, deep learning-based methods [ 26 , 27 ] can directly learn deep-level feature representations from a mass of data to achieve appropriate performance in terms of descriptiveness and robustness. This type of method has proven effective for the registration of indoor and small-scale point clouds; however, it is difficult to apply it to the registration of large-scale MLS and terrestrial laser scanner (TLS) point clouds because of the limitations related to the amount of data and complexity [ 13 ].…”