LAPS: Improving Incremental LiDAR Mapping using Active Pooling and Sampling for Neural Distance Fields

2026-05-14

AI Summary

This work addresses the challenge of catastrophic forgetting in online optimization of neural distance fields for incremental LiDAR mapping, which often leads to degradation of previously reconstructed geometry. To mitigate this issue, the authors propose LAPS, a novel framework that synergistically integrates reliability-based active pooling and uncertainty-guided active sampling. The former dynamically retains critical historical samples to preserve learned structures, while the latter prioritizes under-constrained regions to enhance optimization efficiency. Evaluated on the Blenheim Palace 05 sequence from the Oxford Spires dataset, LAPS significantly outperforms existing passive replay and uniform sampling strategies, achieving a 4.66% improvement in recall and a 3.79% gain in F1 score over PIN-SLAM, with notably enhanced reconstruction completeness.

Abstract

Neural distance fields offer a compact and continuous representation of 3D geometry, making them attractive for incremental LiDAR mapping. However, their online optimization is vulnerable to catastrophic forgetting, where new observations can degrade previously reconstructed geometry. Replay-based training is commonly used to address this issue, but existing methods typically rely on passive replay buffers and uniform sampling, which can waste memory on redundant observations and under-train poorly constrained regions. We propose LAPS, a replay management framework for incremental neural mapping that improves both replay retention and replay allocation during online updates. LAPS combines reliability-based active pooling to retain reliable historical samples under limited memory with uncertainty-guided active sampling to focus optimization on under-constrained regions. Experiments on synthetic and real-world benchmarks show that LAPS consistently improves reconstruction completeness while maintaining competitive geometric accuracy. On Oxford Spires, it improves recall by 4.66 pp and F1-score by 3.79 pp over PIN-SLAM on the Blenheim Palace 05 sequence. We release our open source implementation at: https://github.com/dongjae0107/LAPS.

Problem

• catastrophic forgetting
• incremental LiDAR mapping
• neural distance fields
• replay buffer
• online optimization

Innovation

• active pooling
• active sampling
• neural distance fields
• incremental LiDAR mapping
• catastrophic forgetting