检索结果-内蒙古大学图书馆

International Conference on computer Vision (ICCV)

作者： Ming-Xian Lin Jie Yang He Wang Yu-Kun Lai Rongfei Jia Binqiang Zhao Lin Gao Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Center on Frontiers of Computing Studies Peking University School of Computer Science and Informatics Cardiff University Alibaba Group Zhejiang Lab

ISBN: (纸本)9781665428132

In this work, we tackle the problem of single image-based 3D shape retrieval (IBSR), where we seek to find the most matched shape of a given single 2D image from a shape repository. Most of the existing works learn to embed 2D images and 3D shapes into a common feature space and perform metric learning using a triplet loss. Inspired by the great success in recent contrastive learning works on self-supervised representation learning, we propose a novel IBSR pipeline leveraging contrastive learning. We note that adopting such cross-modal contrastive learning between 2D images and 3D shapes into IBSR tasks is non-trivial and challenging: contrastive learning requires very strong data augmentation in constructed positive pairs to learn the feature invariance, whereas traditional metric learning works do not have this requirement. Moreover, object shape and appearance are entangled in 2D query images, thus making the learning task more difficult than contrasting single-modal data. To mitigate the challenges, we propose to use multi-view grayscale rendered images from the 3D shapes as a shape representation. We then introduce a strong data augmentation technique based on color transfer, which can significantly but naturally change the appearance of the query image, effectively satisfying the need for contrastive learning. Finally, we propose to incorporate a novel category-level contrastive loss that helps distinguish similar objects from different categories, in addition to classic instance-level contrastive loss. Our experiments demonstrate that our approach achieves the best performance on all the three popular IBSR benchmarks, including Pix3D, Stanford Cars, and Comp Cars, outperforming the previous state-of-the-art from 4% - 15% on retrieval accuracy.

关键词： Representation learning Deep learning Three-dimensional displays Shape Image color analysis Pipelines Gray-scale

来源：评论

学校读者我要写书评

暂无评论

Are we ready for a new paradigm shift? A Survey on Visual Deep MLP

arXiv

引用

arXiv 2021年

作者： Liu, Ruiyang Li, Yinghui Tao, Linmi Liang, Dun Zheng, Hai-Tao Department of Computer Science and Technology BNRist Tsinghua University Key Lab of Pervasive Computing Ministry of Education of China Beijing100084 China Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen518055 China

Recently, the proposed deep MLP models have stirred up a lot of interest in the vision community. Historically, the availability of larger datasets combined with increased computing capacity leads to paradigm shifts. This review paper provides detailed discussions on whether MLP can be a new paradigm for computer vision. We compare the intrinsic connections and differences between convolution, self-attention mechanism, and Token-mixing MLP in detail. Advantages and limitations of Token-mixing MLP are provided, followed by careful analysis of recent MLP-like variants, from module design to network architecture, and their applications. In the GPU era, the locally and globally weighted summations are the current mainstreams, represented by the convolution and self-attention mechanism, as well as MLP. We suggest the further development of paradigm to be considered alongside the next-generation computing devices. Copyright © 2021, The Authors. All rights reserved.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Vehicle tracking by detection in UAV aerial video

引用

science China(Information sciences) 2019年第2期62卷 192-194页

作者： Shaohua LIU Suqin WANG Wenhao SHI Haibo LIU Zhaoxin LI Tianlu MAO School of Electronic Engineering University of Beijing Posts and Telecommunications Institute of Electronic and Information Engineering in Guangdong University of Electronic Science and Technology of China School of Control and Computer Engineering North China Electric Power University Beijing Key Lab of Mobile Computing and Pervasive Device Institute of Computing TechnologyChinese Academy of Sciences

With the maturity and popularity of UAV (unmanned aerial vehicle) technology [1–3], UAV video is becoming an effective supplement to the fixed monitoring video [4, 5]. In the aspect of traffic information acquisition, the advantages of UAV are obvious. UAV can fly not only between the buildings, but also on the freeway, and even into the tunnel, showing the unique flexibility and maneuverability. UAV can control its hovering position artificially and has a high angle shot to get more comprehensive and clearer video *** some emergent situation, such as evacuation caused by typhoons and earthquakes, there are lots of countryside areas without fixed road monitoring cameras. UAV is undoubtedly the best choice to

关键词： UAV Vehicle tracking by detection in UAV aerial video

来源：评论

学校读者我要写书评

暂无评论

Recurrent Convolutional Attention Neural Model for Sentiment Classification of short text

Recurrent Convolutional Attention Neural Model for Sentiment...

引用

International Conference on Cutting-edge Technologies in Engineering (ICon-CuTE)

作者： Mohd Usama Belal Ahmad Abhay Pratap Singh Parvez Ahmad Embedded and Pervasive Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Gurukul Kangri vishwavidyalaya Haridwar India School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9781728100012

Nowadays neural attention model doing good in many tasks of natural language processing (NLP). More specifically attention mechanism has been used widely with convolutional neural network (CNN) and recurrent neural network (RNN) for many task of NLP. But the sentiment classification of short text is a challenging task because it contains limited contextual information. Thus, we proposed a new recurrent convolutional attention neural model for sentiment classification of the short text by using attention mechanism with RCNN (recurrent convolutional neural network). In proposed model attention score is calculated by averaging hidden units (feature maps) generated from LSTM (long short-term memory). Then we combined this attention score with recurrent convolution-based encoded text features to get final sentence representation. Here attention will focus on important text features and recurrent convolution makes full use of limited contextual information by processing sentence representation through different windows sizes with specialized recurrent convolution operation. Validation of the proposed model is done through experimentation with three benchmark datasets i.e. MR, SSTl, and SST2. Achieved results exhibit that our model performs better than many existing baselines works on all three datasets.

关键词： convolutional neural nets learning (artificial intelligence) pattern classification recurrent neural nets sentiment analysis recurrent neural nets Pattern recognition texts Sentiment analysis Natural Language Processing NINL gene

来源：评论

学校读者我要写书评

暂无评论

Automatic unpaired shape deformation transfer

Automatic unpaired shape deformation transfer

引用

SIGGRAPH Asia 2018 Technical Papers - International Conference on computer Graphics and Interactive Techniques, SIGGRAPH Asia 2018

作者： Gao, Lin Yang, Jie Qiao, Yi-Ling Xia, Shihong Yang, Jie Qiao, Yi-Ling Lai, Yu-Kun Rosin, Paul L. Xu, Weiwei Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Computer Science and Informatics Cardiff University United Kingdom State Key Lab of CAD and CG Zhejiang University China

ISBN: (纸本)9781450360081

Transferring deformation from a source shape to a target shape is a very useful technique in computer graphics. State-of-the-art deformation transfer methods require either point-wise correspondences between source and target shapes, or pairs of deformed source and target shapes with corresponding deformations. However, in most cases, such correspondences are not available and cannot be reliably established using an automatic algorithm. Therefore, substantial user effort is needed to label the correspondences or to obtain and specify such shape sets. In this work, we propose a novel approach to automatic deformation transfer between two unpaired shape sets without correspondences. 3D deformation is represented in a high-dimensional space. To obtain a more compact and effective representation, two convolutional variational autoencoders are learned to encode source and target shapes to their latent spaces. We exploit a Generative Adversarial Network (GAN) to map deformed source shapes to deformed target shapes, both in the latent spaces, which ensures the obtained shapes from the mapping are indistinguishable from the target shapes. This is still an under-constrained problem, so we further utilize a reverse mapping from target shapes to source shapes and incorporate cycle consistency loss, i.e. applying both mappings should reverse to the input shape. This VAE-Cycle GAN (VC-GAN) architecture is used to build a reliable mapping between shape spaces. Finally, a similarity constraint is employed to ensure the mapping is consistent with visual similarity, achieved by learning a similarity neural network that takes the embedding vectors from the source and target latent spaces and predicts the light field distance between the corresponding shapes. Experimental results show that our fully automatic method is able to obtain high-quality deformation transfer results with unpaired data sets, comparable or better than existing methods where strict correspondences are required.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Correction to “Smart Urban Living: Enabling Emotion-Guided Interaction With Next Generation Sensing Fabric”

引用

IEEE Access 2020年 8卷 96045-96045页

作者： Yingying Jiang Wenjing Xiao Rui Wang Ahmed Barnawi Embedded and Pervasive Computing (EPIC) Laboratory School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Faculty of Computing and IT King Abdulaziz University Jeddah Saudi Arabia

In the above article [1] , there is a mistake in the grant number. The sentence “This work was supported by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under Grant DF-166-611-1441” should be corrected to “This work was supported by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under Grant DF-251-611-1441.”

关键词： Next generation networking Sensors Fabrics Object recognition pervasive computing computer science Licenses

来源：评论

学校读者我要写书评

暂无评论

Job schedulers for Big data processing in Hadoop environment： testing real-life schedulers using benchmark programs

引用

Digital Communications and Networks 2017年第4期3卷 260-273页

作者： Mohd Usama Mengchen Liu Min Chen Embedded and Pervasive Computing(EPIC) Lab School of Computer Science and Technology Huazhong University of Science and Technology wuhon 430074 China

At present, big data is very popular, because it has proved to be much successful in many fields such as social media, E-commerce transactions, etc. Big data describes the tools and technologies needed to capture, manage, store, distribute, and analyze petabyte or larger-sized datasets having different structures with high speed. Big data can be structured, unstructured, or semi structured. Hadoop is an open source framework that is used to process large amounts of data in an inexpensive and efficient way, and job scheduling is a key factor for achieving high performance in big data processing. This paper gives an overview of big data and highlights the problems and challenges in big data. It then highlights Hadoop Distributed File System （HDFS）, Hadoop MapReduce, and various parameters that affect the performance of job scheduling algorithms in big data such as Job Tracker, Task Tracker, Name Node, Data Node, etc. The primary purpose of this paper is to present a comparative study of job scheduling algorithms along with their experimental results in Hadoop environment. In addition, this paper describes the advantages, disadvantages, features, and drawbacks of various Hadoop job schedulers such as FIFO, Fair, capacity, Deadline Constraints, Delay, LATE, Resource Aware, etc, and provides a comparative study among these schedulers.

关键词： Big Data Hadoop MapReduce HDFS Scheduler Classification Locality Benchmark

来源：评论

学校读者我要写书评

暂无评论

Easy transfer learning by exploiting intra-domain structures

arXiv

引用

arXiv 2019年

作者： Wang, Jindong Chen, Yiqiang Yu, Han Huang, Meiyu Yang, Qiang Beijing Key Lab. of Mobile Computing and Pervasive Device Inst. of Computing Tech. CAS School of Computer Science and Engineering Nanyang Technological University Singapore Qian Xuesen Lab. of Space Tech. China Academy of Space Tech. Beijing China Hong Kong University of Science and Technology Hong Kong University of Chinese Academy of Sciences Beijing100190 China

Transfer learning aims at transferring knowledge from a welllabeled domain to a similar but different domain with limited or no labels. Unfortunately, existing learning-based methods often involve intensive model selection and hyperparameter tuning to obtain good results. Moreover, cross-validation is not possible for tuning hyperparameters since there are often no labels in the target domain. This would restrict wide applicability of transfer learning especially in computationallyconstraint devices such as wearables. In this paper, we propose a practically Easy Transfer Learning (EasyTL) approach which requires no model selection and hyperparameter tuning, while achieving competitive performance. By exploiting intra-domain structures, EasyTL is able to learn both nonparametric transfer features and classifiers. Extensive experiments demonstrate that, compared to state-of-the-art traditional and deep methods, EasyTL satisfies the Occam's Razor principle: it is extremely easy to implement and use while achieving comparable or better performance in classification accuracy and much better computational efficiency. Additionally, it is shown that EasyTL can increase the performance of existing transfer feature learning methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Cross-position activity recognition with stratified transfer learning

arXiv

引用

arXiv 2018年

作者： Chen, Yiqiang Wang, Jindong Huang, Meiyu Yu, Han Beijing Key Lab of Mobile Computing and Pervasive Devices Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Qian Xuesen Lab of Space Technology China Academy of Space Technology School of Computer Science and Engineering Nanyang Technological University

Human activity recognition (HAR) aims to recognize the activities of daily living by utilizing the sensors attached to different body parts. The great success of HAR has been achieved by training machine learning models using sufficient labeled activity data. However, when the labeled data from a certain body position (i.e. target domain) is missing, how to leverage the data from other positions (i.e. source domain) to help recognize the activities of this position? This problem can be divided into two steps. Firstly, when there are several source domains available, it is often difficult to select the most similar source domain to the target domain. Secondly, with the selected source domain, we need to perform accurate knowledge transfer between domains in order to recognize the activities on the target domain. Existing methods only learn the global distance between domains while ignoring the local property. In this paper, we propose a Stratified Transfer Learning (STL) framework to perform both source domain selection and activity transfer. STL is based on our proposed Stratified distance to capture the local property of domains. STL consists of two components: 1) Stratified Domain Selection (STL-SDS), which can select the most similar source domain to the target domain;and 2) Stratified Activity Transfer (STL-SAT), which is able to perform accurate knowledge transfer. Extensive experiments on three public activity recognition datasets demonstrate the superiority of STL. Furthermore, we extensively investigate the performance of transfer learning across different degrees of similarities and activity levels between domains. We also discuss the potential applications of STL in other fields of pervasive computing for future research. Copyright © 2018, The Authors. All rights reserved.

关键词： Ubiquitous computing

来源：评论

学校读者我要写书评

暂无评论

Interaction between nearby strangers: Serendipity and playfulness 16

Interaction between nearby strangers: Serendipity and playfu...

引用

9th Nordic Conference on Human-computer Interaction, NordiCHI 2016

作者： Paasovaara, Susanna Lucero, Andrés Porcheron, Martin Paavilainen, Janne Aliakseyeu, Dzmitry Olsson, Thomas Dept of Pervasive Computing Tampere Univ of Technology Tampere Finland Mads Clausen Institute Univ of Southern Denmark Kolding Denmark Mixed Reality Laboratory School of Computer Science University of Nottingham United Kingdom Game Research Lab University of Tampere Tampere Finland Philips Lighting Research Eindhoven Netherlands

ISBN: (纸本)9781450347631

"Nearby strangers" provides an interestingly paradoxical space for interaction design. There are various social norms, cultural practices, and privacy concerns hindering interaction with nearby strangers, but by ignoring them, people constantly miss social opportunities. technology enabling ad-hoc interactions between co-located people has been explored for years in research but real-life applications are still rare. The potential focus areas include increasing awareness of social possibilities;light-weight playful interactions, play and gaming;serendipitous and ad hoc social interaction;anonymous exchange of content;matching interests for various purposes;icebreakers and provocation to interact, and ambient representation of the nearby strangers. This workshop convenes researchers and practitioners to gather and advance the state of research on interactions between nearby strangers. We aim to explore this design space and collaboratively identify new research and design opportunities that novel communication technology creates. © 2016 ACM.

关键词： icebreaking match making proximity-based interaction social encouragement Unfamiliar people collocated interaction

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：