OREN: Octree Residual Network for Real-Time Euclidean Signed Distance Mapping

TL;DR

OREN introduces a hybrid octree-neural network approach for efficient, accurate, and scalable Euclidean signed distance function reconstruction from point clouds, suitable for real-time robotic applications.

Contribution

This work presents OREN, a novel hybrid method combining octree interpolation with neural residuals for improved SDF reconstruction.

Findings

OREN outperforms existing methods in accuracy and efficiency.

It achieves non-truncated Euclidean SDF reconstruction with scalable performance.

OREN is suitable for real-time robotics and computer vision tasks.

Abstract

Reconstructing signed distance functions (SDFs) from point cloud data benefits many robot autonomy capabilities, including localization, mapping, motion planning, and control. Methods that support online and large-scale SDF reconstruction often rely on discrete volumetric data structures, which affects the continuity and differentiability of the SDF estimates. Neural network methods have demonstrated high-fidelity differentiable SDF reconstruction but they tend to be less efficient, experience catastrophic forgetting and memory limitations in large environments, and are often restricted to truncated SDF. This work proposes OREN, a hybrid method that combines an explicit prior from octree interpolation with an implicit residual from neural network regression. Our method achieves non-truncated (Euclidean) SDF reconstruction with computational and memory efficiency comparable to volumetric methods and differentiability and accuracy comparable to neural network methods. Extensive experiments demonstrate that OREN outperforms the state of the art in terms of accuracy and efficiency, providing a scalable solution for downstream tasks in robotics and computer vision.