arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1908.10209cs.LGstat.ML
收藏

混合卷积与综合法有效识别三维目标

Blended Convolution and Synthesis for Efficient Discrimination of 3D Shapes

Australian National University(澳大利亚国立大学);Data61-CSIRO(澳大利亚联邦科学与工业研究组织Data61);Inception Institute of Artificial Intelligence(起源人工智能研究所)

Sameera Ramasinghe, Salman Khan, Nick Barnes, Stephen Gould

摘要:

现有的网络直接学习3D点云上的特征表示以进行形状分析。我们认为,三维点云是高度冗余的,并持有不规则(置换不变)的结构,这使得它很难有效地实现类间的歧视。在本文中,我们提出了一个两方面的解决方案,这个问题是无缝集成在一个单一的“混合卷积和合成”层。这个完全可微的层连续执行两个关键任务。在第一步中,它将输入的3D点云投影到潜在的3D空间中,以合成高度紧凑且更具类间区分性的点云表示。由于3D点云不遵循欧几里德拓扑,因此标准2/3D卷积神经网络提供有限的表示能力。因此,在第二步中,它使用一种新的三维卷积运算符内的单位球($\mathbb{B}^3$)提取有用的体积特征。我们广泛推导公式,以实现我们的新卷积核的平移和旋转。最后,使用所提出的技术,我们提出了一个非常轻的重量,端到端的架构,实现了令人信服的结果,三维形状识别和检索。

英文摘要:

Existing networks directly learn feature representations on 3D point clouds for shape analysis. We argue that 3D point clouds are highly redundant and hold irregular (permutation-invariant) structure, which makes it difficult to achieve inter-class discrimination efficiently. In this paper, we propose a two-faceted solution to this problem that is seamlessly integrated in a single `Blended Convolution and Synthesis' layer. This fully differentiable layer performs two critical tasks in succession. In the first step, it projects the input 3D point clouds into a latent 3D space to synthesize a highly compact and more inter-class discriminative point cloud representation. Since, 3D point clouds do not follow a Euclidean topology, standard 2/3D Convolutional Neural Networks offer limited representation capability. Therefore, in the second step, it uses a novel 3D convolution operator functioning inside the unit ball ($\mathbb{B}^3$) to extract useful volumetric features. We extensively derive formulae to achieve both translation and rotation of our novel convolution kernels. Finally, using the proposed techniques we present an extremely light-weight, end-to-end architecture that achieves compelling results on 3D shape recognition and retrieval.

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。

↑