arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2108.02740cs.CVcs.GR
收藏

WSDesc:弱监督3D局部描述符学习用于点云配准

WSDesc: Weakly Supervised 3D Local Descriptor Learning for Point Cloud Registration

École Polytechnique, IP Paris(巴黎理工学院综合理工学院);City University of Hong Kong(香港城市大学)

Lei Li, Hongbo Fu, Maks Ovsjanikov

摘要:

在这项工作中,我们提出了一种称为WSDesc的新方法,以弱监督的方式学习3D局部描述符,以实现鲁棒的点云配准。我们的工作建立在最近的3D基于CNN的描述符提取器,它利用基于体素的表示来参数化3D点的局部几何形状。我们建议以数据驱动的方式学习最佳支持,而不是在体素化中使用预定义的固定大小的局部支持。为此,我们设计了一种新的可微分体素化层,可以将梯度反向传播到支持大小优化。为了训练提取的描述符,我们提出了一种新的注册损失的基础上偏离刚性的3D变换,和损失是弱监督的先验知识,输入点云有部分重叠,而不需要地面实况对齐信息。通过大量的实验,我们表明,我们学习的描述符产生现有的几何配准基准优越的性能。

英文摘要:

In this work, we present a novel method called WSDesc to learn 3D local descriptors in a weakly supervised manner for robust point cloud registration. Our work builds upon recent 3D CNN-based descriptor extractors, which leverage a voxel-based representation to parameterize local geometry of 3D points. Instead of using a predefined fixed-size local support in voxelization, we propose to learn the optimal support in a data-driven manner. To this end, we design a novel differentiable voxelization layer that can back-propagate the gradient to the support size optimization. To train the extracted descriptors, we propose a novel registration loss based on the deviation from rigidity of 3D transformations, and the loss is weakly supervised by the prior knowledge that the input point clouds have partial overlap, without requiring ground-truth alignment information. Through extensive experiments, we show that our learned descriptors yield superior performance on existing geometric registration benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑