| 基于机器学习的点集匹配算法 |
| |
| 作者姓名： | 唐思琦 韩丛英 郭田德 |
| |
| 作者单位： | 中国科学院大学数学科学学院, 北京 100049;中国科学院大数据挖掘和知识管理重点实验室, 北京 100190 |
| |
| 基金项目： | Supported by the Chinese National Natural Science Foundation (11731013, 11331012, 11571014) |
| |
| 摘 要： | 点集匹配是计算机视觉和模式识别中的重要问题，在目标识别、医学图像配准、姿态估计等方面都得到广泛应用。提出基于机器学习的端对端模型——multi-pointer network（MPN）来解决点集匹配问题。该网络模型利用多标签分类的思想，改进pointer network。以前的模型只输出输入序列的一个元素，而MPN模型选择输入序列中的一组元素作为输出。首先，把点集匹配问题转换为序列问题。这样，网络的输入为顶点的坐标序列，输出为点对之间的对应关系。利用这种方式，可以解决相对于整个空间的平移变换和其他大幅度的刚性变换。实验结果表明，模型也可以被推广解决其他带结构的组合优化问题，如三角剖分等。
|
| 关 键 词： | 多指向型网络 点集匹配 递归神经网络 长短期记忆网络 多标签分类 |
| 收稿时间： | 2018-11-24 |
| 修稿时间： | 2019-01-15 |
Point matching algorithm based on machine learning method |
| |
| Authors: | TANG Siqi HAN Congying GUO Tiande |
| |
| Institution: | School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China;Key Laboratory of Big Data Mining and Knowledge Management of Chinese Academy of Sciences, Beijing 100190, China |
| |
| Abstract: | Point matching is an important issue of computer vision and pattern recognition, and it is widely used in target recognition, medical image, pose estimation, etc. In this study, we propose a novel end-to-end model (multi-pointer network) based on machine learning method to solve this problem. We capitalize on the idea of multi-label classification to ameliorate the pointer network. Instead of outputting a member of input sequence, our model selects a set of input elements as output. Considering matching problem as a sequential manner, our model takes the coordinates of points as input and outputs correspondences directly. Using this new method, we can effectively solve the translation of the whole space and other large-scale rigid transformations. Furthermore, experiment results show that our model can be generalized to other combinatorial optimization problems in which the output is a subset of input, like Delaunay triangulation. |
| |
| Keywords: | multi-pointer network point matching recurrent neural network (RNN) long short-term memory (LSTM) network multi-label classification |
|
| 点击此处可从《》浏览原始摘要信息 |
| 点击此处可从《》下载免费的PDF全文 |
