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 2011.09852cs.LGcs.AI
收藏

快速点特征嵌入的不规则列表MLP

Irregularly Tabulated MLP for Fast Point Feature Embedding

DENSO IT Laboratory(电装IT实验室)

Yusuke Sekikawa, Teppei Suzuki

摘要:

针对测试时点特征嵌入的大幅加速,我们提出了一个新的框架,该框架使用一对多层感知器(MLP)和一个查找表(LUT)将点坐标输入转换为高维特征。与需要数百万点积的MLP实现的PointNet特征嵌入部分相比,在测试时提出的框架不需要这样的矩阵矢量积层,而只需要从列表MLP中查找最近的实体,然后进行插值,定义在基本上不规则排列的3D网格上的离散输入。我们称这个框架为LUTI-MLP:LUT Interpolation ML,它提供了一种训练端到端的不规则列表MLP的方法,该MLP以特定的方式耦合到LUT,而无需在测试时进行任何近似。LUTI-MLP还为测试时李代数$\mathfrak{se}(3)$上的嵌入函数wrt全局姿态坐标的Jacobian计算提供了显著的加速,这可以用于点集配准问题。在使用ModelNet 40进行了广泛的评估之后,我们证实了LUTI-MLP即使有一个小的(例如,$4^3$)格产生与MLP相当的性能,同时实现显著的加速:$100\times$用于嵌入,$12\times$用于近似雅可比矩阵,以及$860\times$用于规范雅可比矩阵。

英文摘要:

Aiming at drastic speedup for point-feature embeddings at test time, we propose a new framework that uses a pair of multi-layer perceptrons (MLP) and a lookup table (LUT) to transform point-coordinate inputs into high-dimensional features. When compared with PointNet's feature embedding part realized by MLP that requires millions of dot products, the proposed framework at test time requires no such layers of matrix-vector products but requires only looking up the nearest entities from the tabulated MLP followed by interpolation, defined over discrete inputs on a 3D lattice that is substantially arranged irregularly. We call this framework LUTI-MLP: LUT Interpolation ML that provides a way to train end-to-end irregularly tabulated MLP coupled to a LUT in a specific manner without the need for any approximation at test time. LUTI-MLP also provides significant speedup for Jacobian computation of the embedding function wrt global pose coordinate on Lie algebra $\mathfrak{se}(3)$ at test time, which could be used for point-set registration problems. After extensive evaluation using the ModelNet40, we confirmed that the LUTI-MLP even with a small (e.g., $4^3$) lattice yields performance comparable to that of the MLP while achieving significant speedup: $100\times$ for the embedding, $12\times$ for the approximate Jacobian, and $860\times$ for the canonical Jacobian.

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。

↑