检索结果-内蒙古大学图书馆

2020 International Symposium on Automation, Information and Computing, ISAIC 2020

作者： Ma, Hao Hu, Dong Li, Yi Jiangsu Province’s Key Lab of Image Procession and Image Communications Nanjing210003 China Education Ministry’s Key Lab of Broadband Wireless Communication and Sensor Network Technology Nanjing210003 China Education Ministry’s Engineering Research Center of Ubiquitous Network and Heath Service Nanjing University of Posts and Telecommunications Nanjing210003 China

This paper presents a joint parallel loop filtering algorithm based on multi-thread load balancing in HEVC decoding, which implements the parallel processing of deblocking filtering (DBF) and sample adaptive compensation (SAO). Because of the diversity of video, the texture of different regions in an image is also different, which leads to various CTU partition methods. Therefore, the number of the boundary to be filtered is greatly different, resulting the computation load among multiple threads unbalanced in parallel processing. To solve this problem, an area division scheme is proposed, which divides the image into multiple areas, and the number of boundaries to be filtered in each area is similar. Then, the mapping relationship table is used to allocate these areas to multiple threads for parallel processing, so as to achieve the load balancing among the filtering threads. Finally, the cache technology is used to combine DBF and SAO to reduce the delay between them and improve the overall parallelism of the loop filter. Experimental results show that the performance of the proposed load balancing joint filtering algorithm is 8.15% higher than the previous scheme. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

The statistic modeling of eye movement viewing S3D images 14th

The statistic modeling of eye movement viewing S3D images

引用

14th International Forum of Digital TV and Wireless Multimedia communication, IFTC 2017

作者： Zhang, Chi Zhou, Jun Zhu, Shoucheng Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Shanghai200240 China Shanghai Key Lab of Digital Media Processing and Transmissions Shanghai Jiao Tong University Shanghai200240 China Shanghai Yanan High School Shanghai200336 China

ISBN: (纸本)9789811081071

Nowadays, more and more families are willing to buy 3D TV to improve their watching experience. Stereo perception produced by watching 3D images or videos brings strong immersive watching experience to users. However, accumulated vision fatigue confuses users a lot after watching 3D TV for a long time. When watching 3D images, controlled by past recognition experience and visual attention mechanism, gaze point of two eyes is changing among different objects which have different depth of field. The eye movement in this changing process is called vergence. Vergence can be defined as movement of our eyes in opposite directions to locate the area of interest on the fovea and accommodation as alteration of the lens to obtain and maintain the area of interest focused on the fovea. So the more frequently the vergence process occurs, the more uncomfortable we feel. We expect to obtain several eye movement patterns, which can be considered as some typical visual attention patterns, by building a top-down recognition and visual attention model and then applying some clustering methods to find them. So we use an eye tracker to record eye movement data and then model it as a bayesian network model. The generative model is based on beta process and we build an Autoregression-HMM model to describe the relationship between latent eye movement patterns and eye movement data. To uncover parameters which represent different eye movement patterns in this model, we use MCMC method to calculate them with iterative computations. In this work, some different latent patterns existed in the sequential eye movement data can be revealed. After analyzing these patterns, we are able to find out some similarities and differences of visual attention models between different people watching the same image or between different images viewed by the same one. These conclusions can help to improve quality of 3D image thus lessening the users’ vision fatigue when watching 3D TV. This work also will con

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Gravity gradient-terrain aided navigation based on particle filter

Gravity gradient-terrain aided navigation based on particle ...

引用

MIPPR 2009 - Remote Sensing and GIS Data processing and Other Applications: 6th International Symposium on Multispectral image processing and Pattern Recognition

作者： Xiong, Ling Ma, Jie Tian, Jin-Wen State Key Lab. for Multi-spectral Information Processing Technologies Key Lab. of Education Ministry for Image Processing and Intelligence Control Huazhong University of Science and Technology Wuhan 430074 China College of Information Science and Engineering Wuhan University of Science and Technology Wuhan 430081 China

ISBN: (纸本)9780819478092

Based on Particle Filter, Gravity Gradient-Terrain aided position technology is proposed in this paper. With the sensitivity of gravity gradient to terrain, the gravity gradient reference map can be computed from the local terrain elevation data. The position can be actualized through matching the real-time measured gravity gradient data to the prepared gravity gradient reference map. The most widely used approximate filtering method is the extended Kaman filter (EKF). EKF is computationally simple but, the convergence of the state estimation for the position is not guaranteed. Particle filter (PF) makes use of the non-linear state and measurement functions, no linearization technology is needed. PF can assure the convergence of the state estimation which follows from the classical results on convergence of Bayesian estimators. Simulations have been done and Particle filter has been shown to be a superior alternative to the EKF in the gravity gradient-terrain matching navigation systems. © 2009 Copyright SPIE - The International Society for Optical Engineering.

关键词： Extended Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Sign Language Recognition using Spiking Neural Networks 2

Sign Language Recognition using Spiking Neural Networks

引用

2nd International Conference on Machine Learning and Data Engineering, ICMLDE 2023

作者： Chaudhari, Pranav Vicente-Sola, Alex Basu, Amlan Manna, Davide L. Kirkland, Paul Caterina, Gaetano Di Department of Information and Communication Technology Manipal Institute of Technology Manipal Academy of Higher Education Manipal576104 India Neuromorphic Sensor Signal Processing Lab Centre for Image and Signal Processing Electrical and Electronic Engineering Department University of Strathclyde Glasgow United Kingdom

In recent years, research in automatic Sign Language Recognition (SLR) has undergone significant progress, serving as a founda-tional base for developing applications that aim to promote the integration of deaf individuals into society. Most of this progress is owed to the recent developments in deep learning. However, the deployment of conventional Artificial Neural Networks (ANNs) can be hindered by their requirements in terms of computational power and energy consumption. Therefore, to improve the ef-ciency of current SLR systems, in this work, we propose the use of the increasingly popular Spiking Neural Networks (SNNs), which, on the one hand, provide more energy-efficient computations than conventional ANNs and, on the other hand, are able to process temporal sequences with simpler architectures thanks to their temporal dynamics. To evaluate our method, we utilize WLASL300, the 300-word (300 classes of signs) dataset from Word-Level American Sign Language, and achieve an improvement in accuracy with the SNN (+2.70%) over the previous state-of-the-art, when working with energy-efficient spiking neurons. Furthermore, we construct a non-spiking version of the same network and evaluate it in a similar manner. Our results demonstrate how the SNN has sparser activations (25% less), thanks to the use of spiking neurons, and therefore can be implemented with a lower power requirement than an ANN version of the same architecture. This work thus demonstrates the possibility of performing SLR in a very effective and efficient way, thus opening up the development of applications that span from the automatic real-time translation of dynamic signs to remote control utilizing sign languages. © 2024 Elsevier B.V.. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Data model driven architecture (DMDA)

Data model driven architecture (DMDA)

引用

4th International Workshop on Advanced Computational Intelligence, IWACI 2011

作者： Chen, Chen Chen, Xueguang Liu, Hong Institute of Systems Engineering Huazhong University of Science and Technology Wuhan 430074 China Key Lab. for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9781612843735

Based on the current development of Model Driven Architecture (MDA) in Enterprise Information System (EIS), the paper proposes a DMDA, a new development architecture to improve EIS development speed and quality. The basic idea of the DMDA is to take advantage of formal description, decomposition and re-use on EIS development process to let computer understand development demands of developers and then consequently generate some business models having nothing to do with specific technical realization, combine with specific technical basic platform to finally generate the EIS meeting actual demands based on data model. The practice proves the architecture is capable of simplifying EIS development difficulty, expediting application and development speed, improve development quality and achieve the rapid development purpose. © 2011 IEEE.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

A probability-based object tracking method

A probability-based object tracking method

引用

4th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2013

作者： Song, Xu Li, Guoqiang Li, Ying Zhang, Yanning Shaanxi Province Key Lab. of Speech and Image Information Processing Northwestern Polytechnical University Xi'an 710129 China Science and Technology on Electro-optic Control Laboratory Louyang 471009 China

ISBN: (纸本)9783642420566

Here we take advantage of the signal recovery power of Compressive Sensing (CS) to significantly reduce the computational complexity brought by the high-dimension image data, then an effective and efficient low-dimensional subspace representation of the object is computing by applying Principal Component Analysis (PCA) to a collection of object observations which are low-dimensional vectors derived from CS. An incremental PCA algorithm is used to update this subspace model for characterizing the object appearance changes. Meanwhile, two distances derived from Probabilistic Principal Component Analysis (PPCA): distance from feature space (DFFS) and distance in feature space (DIFS), are used to describe visual similarity between the learned subspace representation model and candidate targets. Comparing with the traditional used reconstruction error, the sum of two distances: DFFS + DIFS, is more accurate and more robust to noises and partial occlusions. Numerous experiment demonstrate that subspace representation model can handle the situation that target objects experience pose changes, scale changes, significant illumination variation, partial occlusions and so on. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

Epidemic spreading with variant infection rates on scale-free network

引用

6th International Symposium on Neural Networks, ISNN 2009

作者： Hong, Liu Ouyang, Min Mao, Zijun Chen, Xueguang Institute of System Engineering Huazhong University of Science and Technology Wuhan 430074 China Key Lab. for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)3642015123

In this paper, we proposed a susceptible-infected model with variant infection rates because different individuals have different resistance to diseases in different periods of real epidemic events. We consider two cases: Case 1, we know every individual's infection rate to a kind of epidemic, satisfy a type of distribution. Case 2, assume all individuals have same initial infection rates, a susceptible individual's infection rate will be less than the initial rate if he is not infected after limited number of contacts with infected ones. For both two cases, at the time tD, preventive and control measures bring into effects, every individual's infection rate would decrease. We implemented this models on scale-free networks, and found that the epidemic process before the time tD in Case 1 is almost the same as that in Standard SI model if the infection rate in Standard SI model equals the mean infection rate in Case 1. Furthermore, using numerical simulation, we analysis the effects of the parameter tD, and find the bimodal distribution of final infection rates. Finally, we conclude what we get in this paper and give our future direction. © 2009 Springer Berlin Heidelberg.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Adaptive Person-Specific Appearance-Based Gaze Estimation 16th

Adaptive Person-Specific Appearance-Based Gaze Estimation

引用

16th International Forum on Digital TV and Wireless Multimedia communication, IFTC 2019

作者： Zheng, Chuanyang Zhou, Jun Sun, Jun Zhao, Lihua Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Shanghai200240 China Shanghai Key Lab of Digital Media Processing and Transmissions Shanghai Jiao Tong University Shanghai200240 China Children’s Hospital of Shanghai Shanghai200062 China

ISBN: (纸本)9789811533402

Non-invasive gaze estimation from only eye images captured by camera is a challenging problem due to various eye shapes, eye structures and image qualities. Recently, CNN network has been applied to directly regress eye image to gaze direction and obtains good performance. However, generic approaches are susceptible to bias and variance highly relating to different individuals. In this paper, we study the person-specific bias when applying generic methods on new person. And we introduce a novel appearance-based deep neural network integrating meta-learning to reduce the person-specific bias. Given only a few person-specific calibration images collected in normal calibration process, our model adapts quickly to test person and predicts more accurate gaze directions. Experiments on public MPIIGaze dataset and Eyediap dataset show our approach has achieved competitive accuracy to current state-of-the-art methods and are able to alleviate person-specific bias problem. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Error concealment techniques for video transmission over error-prone channels: a survey

引用

Journal of Computational Information Systems 2012年第21期8卷 8807-8818页

作者： Cui, Ziguan Gan, Zongliang Zhan, Xuefeng Zhu, Xiuchang Image Processing and Image Communication Lab Nanjing University of Posts and Telecommunications Nanjing 210003 China Key Lab of 'Broadband Wireless Communication and Sensor Network Technology' Nanjing University of Posts and Telecommunications Ministry of Education China Institute of Physics and Communication Electronics Jiangxi Normal University Nanchang 330027 China

Efficient video transmission over unreliable channels may encounter huge challenge due to unavoidable bit error or packets loss. Error concealment (EC) techniques at the decoder side have been developed to recover the damaged regions utilizing spatial or temporal redundant information without changing the encoder structure or adding extra bandwidth. In this work the classic EC techniques and their developments are first reviewed, high-level semantics based EC schemes are also surveyed, and then the emphasis is focused on new EC features introduced by H.264/AVC. Finally, the challenges and future development directions in EC for advanced video coding schemes such as scalable video coding (SVC), multiple description coding (MDC), multi-view video coding (MVC) and stereo video coding are prospected, and future research directions are also indicated according to the current research status and existent problems. © 2012 Binary Information Press.

关键词： Scalable video coding

来源：评论

学校读者我要写书评

暂无评论

Multi-agent middleware for the integration of mobile supply chain

引用

Journal of Computers 2011年第7期6卷 1469-1476页

作者： Zhao, Weidong Wu, Haifeng Dai, Weihui Li, Xuan Yu, Fei Xu, Chen School of Software Fudan University Shanghai China Jiangsu Provincial Key Lab of Image Processing and Image Communications Nanjing University of Posts and Telecommunications Nanjing China

With the development of mobile network and communication technology, traditional supply chain management is gradually updating to mobile supply chain management, and multi-agent technology has been considered as a very efficient method for supply chain management in mobile environment. After considering the characteristics and limitations of mobile supply chain, and according to the requirements in the manufacturing domain, this paper designs a multi-agent middleware for mobile supply chain management, aiming to solve integration problems and shield off network and terminal heterogeneity in mobile environment, which can achieve mobile supply chain dynamic integration and then improve the efficiency of supply chain management. Its feasibility and efficiency was demonstrated by experiments. © 2011 ACADEMY PUBLISHER.

关键词： Middleware

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：