An Interpretable Joint Nonnegative Matrix Factorization-Based Point Cloud Distance Measure
An Interpretable Joint Nonnegative Matrix Factorization-Based Point Cloud Distance Measure
复制标题
DOI：
10.1109/ciss56502.2023.10089765
复制
发表时间：
2022-07
期刊：
影响因子：
--
通讯作者：
Hannah Friedman;Amani Maina-Kilaas;Julianna Schalkwyk;Hina Ahmed;Jamie Haddock
中科院分区：
文献类型：
--
作者：
Hannah Friedman;Amani Maina-Kilaas;Julianna Schalkwyk;Hina Ahmed;Jamie Haddock
本文提出了一种确定数据集或点云的共享特征和测量点云之间距离的新方法。我们的方法使用两个数据矩阵$X_{1}, X_{2}$的联合因式分解成非负矩阵$X_{1}=AS_{1}, X_{2}=AS_{2}$，以得出一个相似性度量，该度量确定共享基${a}$近似$X_{1}, X_{2}$的程度。我们还提出了一种基于该方法和学习到的分解的点云距离度量。我们的方法揭示了图像和文本数据的结构差异。潜在的应用包括分类、检测剽窃或其他操作、数据去噪和迁移学习。
In this paper, we propose a new method for deter-mining shared features of and measuring the distance between data sets or point clouds. Our approach uses the joint factorization of two data matrices $X_{1}, X_{2}$ into non-negative matrices $X_{1}=AS_{1}, X_{2}=AS_{2}$ to derive a similarity measure that determines how well the shared basis ${A}$ approximates $X_{1}, X_{2}$. We also propose a point cloud distance measure built upon this method and the learned factoriI zation. Our method reveals structural differences in both image and text data. Potential applications include classification, detecting plagiarism or other manipulation, data denoising, and transfer learning.