Deep Learning of Warping Functions for Shape Analysis.
Deep Learning of Warping Functions for Shape Analysis.
复制标题
DOI：
10.1109/cvprw50498.2020.00441
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Joshi SH
中科院分区：
文献类型：
--
作者：
Nunez E;Joshi SH
函数与曲线形状之间的率不变或重参数化不变匹配是计算机视觉和医学成像中的一个重要问题。通常，使用诸如动态时间规整或动态编程之类的方法进行匹配的计算成本对于大型数据集是过高的。在这里，我们提出了一种基于深度神经网络的方法，用于从由大量最佳匹配组成的训练数据中学习扭曲函数，并使用它来预测最佳的同构扭曲函数。结果显示，在来自ETH-80数据集的凹凸函数和二维曲线的合成数据集上的预测性能以及计算成本的显着降低。
Rate-invariant or reparameterization-invariant matching between functions and shapes of curves, respectively, is an important problem in computer vision and medical imaging. Often, the computational cost of matching using approaches such as dynamic time warping or dynamic programming is prohibitive for large datasets. Here, we propose a deep neural-network-based approach for learning the warping functions from training data consisting of a large number of optimal matches, and use it to predict optimal diffeomorphic warping functions. Results show prediction performance on a synthetic dataset of bump functions and two-dimensional curves from the ETH-80 dataset as well as a significant reduction in computational cost.