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 2407.09167cs.AIcs.LG
收藏

用于点云装配的 SE(3)-双等变 Transformer

SE(3)-bi-equivariant Transformers for Point Cloud Assembly

University of Gothenburg(哥德堡大学);Chalmers University of Technology(查尔姆斯理工大学)

Ziming Wang, Rebecka Jörnsten

AI总结:

针对非重叠且初始位置任意的点云装配问题,提出基于 SE(3)-双等变先验的 BITR,通过 $SE(3) \ imes SE(3)$-transformer 提取特征并投影至 SE(3),兼具尺度与交换等变性及实验有效性。

摘要:

给定一对点云,装配的目标是恢复一个刚性变换,将一个点云与另一个点云对齐。这项任务具有挑战性,因为点云可能是非重叠的,并且它们可能具有任意的初始位置。为了解决这些困难,我们提出了一种称为 SE(3)-双等变 Transformer(BITR)的方法,该方法基于该任务的 SE(3)-双等变性先验:它保证当输入受到刚性扰动时,输出将相应地发生变换。得益于其等变性,BITR 不仅能够处理非重叠 PC,还能保证对初始位置的鲁棒性。具体而言,BITR 首先使用一种新颖的 $SE(3) \ imes SE(3)$-transformer 提取输入特征,然后将学习到的特征投影到群 SE(3) 上作为输出。此外,我们从理论上证明,交换等变性和尺度等变性可以被纳入 BITR,因此它进一步保证了在缩放和交换输入时性能的稳定性。我们通过实验证明了 BITR 在实际任务中的有效性。

英文摘要:

Given a pair of point clouds, the goal of assembly is to recover a rigid transformation that aligns one point cloud to the other. This task is challenging because the point clouds may be non-overlapped, and they may have arbitrary initial positions. To address these difficulties, we propose a method, called SE(3)-bi-equivariant transformer (BITR), based on the SE(3)-bi-equivariance prior of the task: it guarantees that when the inputs are rigidly perturbed, the output will transform accordingly. Due to its equivariance property, BITR can not only handle non-overlapped PCs, but also guarantee robustness against initial positions. Specifically, BITR first extracts features of the inputs using a novel $SE(3) \times SE(3)$-transformer, and then projects the learned feature to group SE(3) as the output. Moreover, we theoretically show that swap and scale equivariances can be incorporated into BITR, thus it further guarantees stable performance under scaling and swapping the inputs. We experimentally show the effectiveness of BITR in practical tasks.

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。

↑