GERA: Geometric Embedding for Efficient Point Registration Analysis

TL;DR

This paper introduces GERA, a pure MLP-based point cloud registration method that constructs geometric information offline, reducing computational costs and improving generalization for resource-limited applications.

Contribution

The paper presents a novel MLP-based registration network that replaces complex modules with offline geometric encoding, enhancing efficiency and stability.

Findings

Reduces inference time and memory usage significantly.

Improves generalization and stability over traditional methods.

Achieves competitive accuracy with lower resource demands.

Abstract

Point cloud registration aims to provide estimated transformations to align point clouds, which plays a crucial role in pose estimation of various navigation systems, such as surgical guidance systems and autonomous vehicles. Despite the impressive performance of recent models on benchmark datasets, many rely on complex modules like KPConv and Transformers, which impose significant computational and memory demands. These requirements hinder their practical application, particularly in resource-constrained environments such as mobile robotics. In this paper, we propose a novel point cloud registration network that leverages a pure MLP architecture, constructing geometric information offline. This approach eliminates the computational and memory burdens associated with traditional complex feature extractors and significantly reduces inference time and resource consumption. Our method is the first to replace 3D coordinate inputs with offline-constructed geometric encoding, improving generalization and stability, as demonstrated by Maximum Mean Discrepancy (MMD) comparisons. This efficient and accurate geometric representation marks a significant advancement in point cloud analysis, particularly for applications requiring fast and reliability.