Search

Your search keyword '"ZHANG, Mingming"' showing total 6,769 results
Start Over Author "ZHANG, Mingming"
6,769 results on '"ZHANG, Mingming"'

176. IMU Data Processing For Inertial Aided Navigation: A Recurrent Neural Network Based Approach

Author

Zhang, Ming, Zhang, Mingming, Chen, Yiming, and Li, Mingyang

Subjects
Computer Science - Robotics, Computer Science - Computer Vision and Pattern Recognition
Abstract

In this work, we propose a novel method for performing inertial aided navigation, by using deep neural networks (DNNs). To date, most DNN inertial navigation methods focus on the task of inertial odometry, by taking gyroscope and accelerometer readings as input and regressing for integrated IMU poses (i.e., position and orientation). While this design has been successfully applied on a number of applications, it is not of theoretical performance guarantee unless patterned motion is involved. This inevitably leads to significantly reduced accuracy and robustness in certain use cases. To solve this problem, we design a framework to compute observable IMU integration terms using DNNs, followed by the numerical pose integration and sensor fusion to achieve the performance gain. Specifically, we perform detailed analysis on the motion terms in IMU kinematic equations, propose a dedicated network design, loss functions, and training strategies for the IMU data processing, and conduct extensive experiments. The results show that our method is generally applicable and outperforms both traditional and DNN methods by wide margins., Comment: IEEE International Conference on Robotics and Automation (ICRA), Xi'an, China, 2021

Published
2021

178. Semi-supervised Hyperspectral Image Classification with Graph Clustering Convolutional Networks

Author

Zeng, Hao, Liu, Qingjie, Zhang, Mingming, Han, Xiaoqing, and Wang, Yunhong

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Hyperspectral image classification (HIC) is an important but challenging task, and a problem that limits the algorithmic development in this field is that the ground truths of hyperspectral images (HSIs) are extremely hard to obtain. Recently a handful of HIC methods are developed based on the graph convolution networks (GCNs), which effectively relieves the scarcity of labeled data for deep learning based HIC methods. To further lift the classification performance, in this work we propose a graph convolution network (GCN) based framework for HSI classification that uses two clustering operations to better exploit multi-hop node correlations and also effectively reduce graph size. In particular, we first cluster the pixels with similar spectral features into a superpixel and build the graph based on the superpixels of the input HSI. Then instead of performing convolution over this superpixel graph, we further partition it into several sub-graphs by pruning the edges with weak weights, so as to strengthen the correlations of nodes with high similarity. This second round of clustering also further reduces the graph size, thus reducing the computation burden of graph convolution. Experimental results on three widely used benchmark datasets well prove the effectiveness of our proposed framework.

Published
2020

179. A Novel Transformation Approach of Shared-link Coded Caching Schemes for Multiaccess Networks

Author

Cheng, Minquan, Wan, Kai, Liang, Dequan, Zhang, Mingming, and Caire, Giuseppe

Subjects
Computer Science - Information Theory
Abstract

This paper considers the multiaccess coded caching systems formulated by Hachem et al., including a central server containing $N$ files connected to $K$ cache-less users through an error-free shared link, and $K$ cache-nodes, each equipped with a cache memory size of $M$ files. Each user has access to $L$ neighbouring cache-nodes with a cyclic wrap-around topology. The coded caching scheme proposed by Hachem et al. suffers from the case that $L$ does not divide $K$, where the needed number of transmissions (a.k.a. load) is at most four times the load expression for the case where $L$ divides $K$. Our main contribution is to propose a novel {\it transformation} approach to smartly extend the schemes satisfying some conditions for the well known shared-link caching systems to the multiaccess caching systems. Then we can get many coded caching schemes with different subpacketizations for multiaccess coded caching system. These resulting schemes have the maximum local caching gain (i.e., the cached contents stored at any $L$ neighbouring cache-nodes are different such that the number of retrieval packets by each user from the connected cache-nodes is maximal) and the same coded caching gain as the original schemes. Applying the transformation approach to the well-known shared-link coded caching scheme proposed by Maddah-Ali and Niesen, we obtain a new multiaccess coded caching scheme that achieves the same load as the scheme of Hachem et al. but for any system parameters. Under the constraint of the cache placement used in this new multiaccess coded caching scheme, our delivery strategy is approximately optimal when $K$ is sufficiently large. Finally, we also show that the transmission load of the proposed scheme can be further reduced by compressing the multicast message., Comment: 19 pages

Published
2020

180. Research on Patent Valuation Based on Social Network Analysis Method

Author

You, Xuemei, Zhang, Mingming, Yu, Yangli, Li, Kan, Editor-in-Chief, Li, Qingyong, Associate Editor, Fournier-Viger, Philippe, Series Editor, Hong, Wei-Chiang, Series Editor, Liang, Xun, Series Editor, Wang, Long, Series Editor, Xu, Xuesong, Series Editor, Guan, Guiyun, editor, Qu, Bo, editor, and Zhou, Ding, editor

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results