DB-TSDF: Directional Bitmask-based Truncated Signed Distance Fields for Efficient Volumetric Mapping

2025-09-24

AI Summary

In pure-CPU environments, 3D volumetric mapping using Truncated Signed Distance Functions (TSDF) suffers from low efficiency, with processing time scaling significantly with voxel resolution. Method: This paper proposes a directional bit-mask-based incremental TSDF fusion framework. It encodes voxel update directions via bit masks to enable constant-time voxel access and atomic fusion from point clouds, integrated with lightweight voxel grid management and multithreaded parallelization. Contribution/Results: The approach eliminates resolution-dependent latency per frame. Experiments on public LiDAR datasets demonstrate real-time, high-resolution, dense, and geometrically consistent TSDF mapping without GPU acceleration. Its accuracy matches state-of-the-art GPU-based methods, while inference speed surpasses all existing CPU-only approaches. This work is the first to empirically validate the feasibility and practicality of real-time, high-accuracy TSDF reconstruction entirely on CPU.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

This paper presents a high-efficiency, CPU-only volumetric mapping framework based on a Truncated Signed Distance Field (TSDF). The system incrementally fuses raw LiDAR point-cloud data into a voxel grid using a directional bitmask-based integration scheme, producing dense and consistent TSDF representations suitable for real-time 3D reconstruction. A key feature of the approach is that the processing time per point-cloud remains constant, regardless of the voxel grid resolution, enabling high resolution mapping without sacrificing runtime performance. In contrast to most recent TSDF/ESDF methods that rely on GPU acceleration, our method operates entirely on CPU, achieving competitive results in speed. Experiments on real-world open datasets demonstrate that the generated maps attain accuracy on par with contemporary mapping techniques.

Problem

• Efficiently fuses LiDAR point-clouds into volumetric maps
• Maintains constant processing time regardless of grid resolution
• Achieves real-time 3D reconstruction using only CPU processing

Innovation

• Directional bitmask-based TSDF integration scheme
• Constant-time processing independent of voxel resolution
• CPU-only implementation achieving real-time performance