通过重叠区域采样，高效降低点云配准内存占用。
Memory-Efficient Point Cloud Registration via Overlapping Region Sampling
- 聚焦重叠区域密集采样，结合kNN压缩机制
- 3DMatch上达94%召回率，内存减少33%
- 适合资源受限环境下的大规模点云配准
深度学习提升了3D点云配准性能，但显著增加了GPU内存消耗，通常需预采样，牺牲精度。本文提出一种重叠区域采样方法，在保持精度的同时降低内存使用。该方法估计点云重叠区域，并在其中进行密集采样，结合基于k近邻（kNN）的点压缩机制，融合多层感知机（MLP）与Transformer架构。在3DMatch和3DLoMatch数据集上的评估表明，本方法在注册召回率上优于其他采样策略，尤其在低内存条件下优势更明显。在3DMatch上实现94%召回率，内存使用减少33%，在3DLoMatch上表现更优。该方法使资源受限环境下高效的大规模点云配准成为可能，显著降低内存需求同时保持高精度。
原文摘要 · Abstract (English)
Recent advances in deep learning have improved 3D point cloud registration but increased graphics processing unit (GPU) memory usage, often requiring preliminary sampling that reduces accuracy. We propose an overlapping region sampling method to reduce memory usage while maintaining accuracy. Our approach estimates the overlapping region and intensively samples from it, using a k-nearest-neighbor (kNN) based point compression mechanism with multi layer perceptron (MLP) and transformer architectures. Evaluations on 3DMatch and 3DLoMatch datasets show our method outperforms other sampling methods in registration recall, especially at lower GPU memory levels. For 3DMatch, we achieve 94% recall with 33% reduced memory usage, with greater advantages in 3DLoMatch. Our method enables efficient large-scale point cloud registration in resource-constrained environments, maintaining high accuracy while significantly reducing memory requirements.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。