报告题目：Learned Binary Spectral Shape Descriptor for 3D Shape Correspondence （用于三维形状匹配的基于学习的二值化谱形状描述子）
报告摘要：Recently, the local shape descriptor based 3D shape correspondence approaches have been widely studied, where the local shape descriptor is a real-valued vector to characterize geometric structures of shapes. Different from these real-valued local shape descriptors, in this work, we propose to learn a novel binary spectral shape descriptor with the deep neural network for 3D shape correspondence. The binary spectral shape descriptor can require less storage space and enable fast matching. First, based on the eigenvectors of the Laplace-Beltrami operator, we construct a neural network to form a nonlinear spectral representation. Then, for the defined positive and negative points, we can train the constructed neural network by minimizing the errors between the outputs and their corresponding binary descriptors, minimizing the variations of the outputs of the positive points and maximizing the variations of the outputs of the negative points simultaneously. Finally, we binarize the output of the neural network to form a binary spectral shape descriptor. The experimental results demonstrate the effectiveness of the proposed binary shape descriptor for the shape correspondence task.。
个人简介： Jin Xie received his Ph.D. degree from the Department of Computing, The Hong Kong Polytechnic University. He was a postdoctoral associate at New York University Abu Dhabi. Now he is a research scientist at New York University Abu Dhabi. His research interests include computer vision and machine learning. He is currently focusing on 3D computer vision with convex optimization and deep learning methods.