Explore the Influence of Shallow Information on Point Cloud Registration

Feature extraction is a key step for deep-learning-based point cloud registration. In the correspondence-free point cloud registration task, the previous work commonly aggregates deep information for global feature extraction and numerous shallow information which is positive to point cloud registration will be ignored with the deepening of the neural network. Shallow information tends to represent the structural information of the point cloud, while deep information tends to represent the semantic information of the point cloud. In addition, fusing information of different dimensions is conducive to making full use of shallow information. Inspired by this, we verify shallow information in the middle layers can bring a positive impact on the point cloud registration task. We design various architectures to combine shallow information and deep information to extract global features for point cloud registration. Experimental results on the ModelNet40 dataset illustrate that feature extractors that incorporate shallow information will bring positive performance.

Automated summary

Feature extraction is crucial for deep-learning-based point cloud registration. Previous work often neglects shallow information, which is important for point cloud registration, as the neural network deepens. Shallow information represents the structure of the point cloud, while deep information represents its semantics. We demonstrate that incorporating shallow information in feature extractors positively impacts point cloud registration, as evidenced by experimental results on the ModelNet40 dataset.