IEEE Trans Image Process. 2023;32:2734-2748. doi: 10.1109/TIP.2023.3272821. Epub 2023 May 19.
Point cloud shape correspondence aims at accurately mapping one point cloud to another point cloud with various 3D shapes. Since point clouds are usually sparse, disordered, irregular, and with diverse shapes, it is challenging to learn consistent point cloud representations and achieve the accurate matching of different point cloud shapes. To address the above issues, we propose a Hierarchical Shape-consistent TRansformer for unsupervised point cloud shape correspondence (HSTR), including a multi-receptive-field point representation encoder and a shape-consistent constrained module in a unified architecture. The proposed HSTR enjoys several merits. In the multi-receptive-field point representation encoder, we set progressively larger receptive fields in different blocks to simultaneously consider the local structure and the long-range context. In the shape-consistent constrained module, we design two novel shape selective whitening losses, which can complement each other to achieve suppression of features sensitive to shape change. Extensive experimental results on four standard benchmarks demonstrate the superiority and generalization ability of our approach to existing methods at the similar model scale, and our method achieves the new state-of-the-art results.
点云形状对应旨在将一个点云准确地映射到另一个具有各种 3D 形状的点云上。由于点云通常是稀疏的、无序的、不规则的，并且形状多样，因此学习一致的点云表示并实现不同点云形状的准确匹配具有挑战性。为了解决上述问题，我们提出了一种分层形状一致的 Transformer 用于无监督点云形状对应（HSTR），包括多感受野点表示编码器和统一架构中的形状一致约束模块。所提出的 HSTR 具有几个优点。在多感受野点表示编码器中，我们在不同的块中设置逐渐增大的感受野，以同时考虑局部结构和长程上下文。在形状一致的约束模块中，我们设计了两种新颖的形状选择白化损失，它们可以相互补充，以实现对形状变化敏感的特征的抑制。在四个标准基准上的广泛实验结果表明，在类似的模型规模下，我们的方法相对于现有方法具有优越性和泛化能力，并且我们的方法达到了新的最先进的结果。
