Self-Supervised Point Cloud Registration with Deep Versatile Descriptors

TL;DR

This paper introduces a self-supervised point cloud registration method that combines global and local descriptors, improving accuracy and robustness over existing methods, and is suitable for intelligent vehicle applications.

Contribution

The paper proposes a novel self-supervised registration approach using combined global and local descriptors, enhancing feature representation and registration performance.

Findings

Outperforms existing unsupervised registration methods.

Surpasses some supervised methods in accuracy.

Demonstrates robustness and efficiency for intelligent vehicle use.

Abstract

As a fundamental yet challenging problem in intelligent transportation systems, point cloud registration attracts vast attention and has been attained with various deep learning-based algorithms. The unsupervised registration algorithms take advantage of deep neural network-enabled novel representation learning while requiring no human annotations, making them applicable to industrial applications. However, unsupervised methods mainly depend on global descriptors, which ignore the high-level representations of local geometries. In this paper, we propose to jointly use both global and local descriptors to register point clouds in a self-supervised manner, which is motivated by a critical observation that all local geometries of point clouds are transformed consistently under the same transformation. Therefore, local geometries can be employed to enhance the representation ability of the feature extraction module. Moreover, the proposed local descriptor is flexible and can be integrated into most existing registration methods and improve their performance. Besides, we also utilize point cloud reconstruction and normal estimation to enhance the transformation awareness of global and local descriptors. Lastly, extensive experimental results on one synthetic and three real-world datasets demonstrate that our method outperforms existing state-of-art unsupervised registration methods and even surpasses supervised ones in some cases. Robustness and computational efficiency evaluations also indicate that the proposed method applies to intelligent vehicles.