学习实例感知对应关系以实现杂乱场景中鲁棒的多实例点云配准
Learning Instance-Aware Correspondences for Robust Multi-Instance Point Cloud Registration in Cluttered Scenes
AI总结：
针对杂乱场景多实例点云配准中对应关系易受背景和其他实例干扰的问题，提出由粗到细的 MIRETR，利用实例掩码提取可靠对应并进行候选优化，在 ROBI 上 F1 提升 16.6 分。
摘要：
多实例点云配准用于估计场景点云中某个模型点云的多个实例的位姿。提取准确的点对应关系是该问题的核心。现有方法通常将场景点云视为一个整体，忽视了实例之间的分离。因此，点特征很容易受到来自背景或不同实例的其他点污染，从而产生忽视独立实例的不准确对应关系，在杂乱场景中尤其如此。本文提出 MIRETR（Multi-Instance REgistration TRansformer，多实例配准 Transformer），这是一种由粗到细的实例感知对应关系提取方法。在粗层级，它联合学习实例感知超点特征，并预测逐实例掩码。借助实例掩码，所关注实例外部带来的影响被降至最低，从而能够提取高度可靠的超点对应关系。随后，在细层级根据实例掩码将超点对应关系扩展为实例候选。最后，本文设计了一种高效的候选选择与优化算法，以获得最终配准结果。在三个公开基准上的大量实验证明了该方法的有效性。特别是在具有挑战性的 ROBI 基准上，MIRETR 的 F1 分数优于现有最优方法 16.6 分。代码和模型可在 https://github.com/zhiyuanYU134/MIRETR 获取。
英文摘要：
Multi-instance point cloud registration estimates the poses of multiple instances of a model point cloud in a scene point cloud. Extracting accurate point correspondence is to the center of the problem. Existing approaches usually treat the scene point cloud as a whole, overlooking the separation of instances. Therefore, point features could be easily polluted by other points from the background or different instances, leading to inaccurate correspondences oblivious to separate instances, especially in cluttered scenes. In this work, we propose MIRETR, Multi-Instance REgistration TRansformer, a coarse-to-fine approach to the extraction of instance-aware correspondences. At the coarse level, it jointly learns instance-aware superpoint features and predicts per-instance masks. With instance masks, the influence from outside of the instance being concerned is minimized, such that highly reliable superpoint correspondences can be extracted. The superpoint correspondences are then extended to instance candidates at the fine level according to the instance masks. At last, an efficient candidate selection and refinement algorithm is devised to obtain the final registrations. Extensive experiments on three public benchmarks demonstrate the efficacy of our approach. In particular, MIRETR outperforms the state of the arts by 16.6 points on F1 score on the challenging ROBI benchmark. Code and models are available at https://github.com/zhiyuanYU134/MIRETR.
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。
