检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Wang, Xin Lv, Rongrong Zhao, Yang Yang, Tangwen Ruan, Qiuqi Institute of Information Science School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China Guangdong Key Laboratory of Intelligent Information processing Shenzhen University ShenzhenGuangdong518060 China

Crowd counting aims to predict the number of people and generate the density map in the image. There are many challenges, including varying head scales, the diversity of crowd distribution across images and cluttered backgrounds. In this paper, we propose a multi-scale context aggregation network (MSCANet) based on single column encoder-decoder architecture for crowd counting, which consists of an encoder based on dense context-aware module (DCAM) and a hierarchical attention-guided decoder. To handle the issue of scale variation, we construct the DCAM to aggregate multi-scale contextual information through densely connecting the dilated convolution with varying receptive fields. The proposed DCAM can capture rich contextual information of crowd areas due to its long-range receptive fields and dense scale sampling. Moreover, to suppress the background noise and generate a high-quality density map, we adopt a hierarchical attention-guided mechanism in the decoder. This helps to integrate more useful spatial information from shallow feature maps of the encoder by introducing multiple supervision based on semantic attention module (SAM). Extensive experiments demonstrate that the proposed approach achieves better performance than other similar state-of-the-art methods on three challenging benchmark datasets for crowd counting. The code is available at https://***/KingMV/MSCANet. © 2021, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based intelligent network management for 5g and beyond 3

Blockchain-based intelligent network management for 5g and b...

引用

3rd International Conference on Advanced information and Communications Technologies, AICT 2019

作者： Maksymyuk, Taras Gazda, Juraj Han, Longzhe Jo, Minho Department of Telecommunications Lviv Polytechnic National University Lviv Ukraine Department of Computers and Informatics Technical University of Kosice Kosice Slovakia Jiangxi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology Nanchang China Department of Computer Convergence Software Korea University Sejong Metropolitan City Korea Republic of

ISBN: (纸本)9781728123998

The concept of spectrum and infrastructure sharing appeared to solve the problem by allowing operators to cooperate between each other. However, the current architecture of mobile network, which is mostly based on the centralized management entities is not suitable for the spectrum and infrastructure sharing due to the complexity of billing procedure. Moreover, the operator oriented mobile network is slowing the overall technological development, because operators must balance between capital expenditures and profit margins. Such situation negatively affects users, because they are either experience lower quality of service or forced to pay higher price. In this paper, we propose a new intelligent network architecture, which leverage the blockchain technology to handle relationship between operators and users based on smart contracts. The new unlicensed spectrum sharing algorithm between operators using the virtual cryptocurrency has been developed based on game theory. Simulation results show that proposed algorithm achieves Nash equilibrium between operators within few seconds. © 2019 IEEE.

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Language-Image Pre-Training 24

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Lang...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Li, Yudong Hou, Xianxu Dezhi, Zheng Shen, Linlin Zhao, Zhe School of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Tencent AI Lab Beijing China

ISBN: (纸本)9798400706868

While significant progress has been made in multi-modal learning driven by large-scale image-text datasets, there is still a noticeable gap in the availability of such datasets within the facial domain. To facilitate and advance the field of facial representation learning, we present FLIP-80M, a large-scale visual-linguistic dataset comprising over 80 million face images paired with text descriptions. FLIP-80M is constructed by leveraging the large openly available image-text-pair dataset LAION-5B and a mixed-method approach to filter face-related pairs from both visual and linguistic perspectives. Our curation process involves face detection, face caption classification, text de-noising, and synthesis-based image augmentation. As a result, FLIP-80M stands as the largest face-text dataset to date. To evaluate the potential of our dataset, we fine-tune the CLIP model using the proposed FLIP-80M, to create FLIP (Facial Language-Image Pretraining) and assess its representation capabilities across various downstream tasks. Our experiments demonstrate that our FLIP model achieves state-of-the-art results in a range of face analysis tasks, including face parsing, face alignment, and face attribute classification. The dataset and models are available at https://***/ydli-ai/FLIP. © 2024 ACM.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

An efficient sparse representation based classification for undersampled face recognition

An efficient sparse representation based classification for ...

引用

Congress on Evolutionary Computation

作者： Zizhu Fan Lipan Kang Zuoyong Li School of Basic Science East China Jiaotong University Nanchang China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Fuzhou China

ISBN: (纸本)9781509006243

The typical sparse representation for classification (SRC) can obtain desirable recognition result when the training samples in each class are sufficient. Nevertheless, if the training sample set is small scale, i.e., each class has a few training samples, even single sample, the traditional SRC cannot perform well. Although one of the variants of the traditional SRC, the extended SRC(ESRC), can effectively address the above small-scale training set (SSTS) problem, its computational efficiency is very low and consequently constrains the application of the ESRC algorithm. In order to improve the computational efficiency of the ESRC algorithm, we propose a new algorithm based on coordinate descent scheme in this work. Our proposed algorithm is referred as to the fast extended SRC (FESRC) algorithm. Experiments on popular face datasets show that the FESRC algorithm can obtain the high computational efficiency without significantly degrading the recognition results.

关键词： Algorithm design and analysis Training Minimization Computational efficiency Classification algorithms Face Face recognition

来源：评论

学校读者我要写书评

暂无评论

Image saliency detection via multi-feature and manifold-space ranking 2021

Image saliency detection via multi-feature and manifold-spac...

引用

3rd Asia Pacific information Technology Conference, APIT 2021

作者： Li, Xiaoli Zhao, Huaici Liu, Yunpeng Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China Key Laboratory of Opto-Electronic Information Processing Shenyang110016 China The Key Lab of Image Understanding and Computer Vision Shenyang110016 China Shenyang Jianzhu University Shenyang110168 China

ISBN: (纸本)9781450388108

In this paper, we propose an image saliency detection method by using multi-feature and manifold-space ranking. Basically, the proposed method extracts the color-histogram feature to obtain the fine information of the image, and the color-mean feature to obtain the coarse information respectively. To further improve the detection accuracy of the feature correlation between different image units, a manifold-space ranking method is used to calculate saliency values of image units to construct a saliency map on each feature-space. Specifically, we fuse the two saliency maps to obtain the final saliency map. Extensive experiments demonstrate that the proposed method not only outperforms the other methods, but also improves the accuracy and robustness of the saliency detection. © 2021 ACM.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Passive Localization of WiFi Client Based on Location Fingerprint

Passive Localization of WiFi Client Based on Location Finger...

引用

2016 3rd International Conference on Engineering Technology and Application（ICETA 2016）

作者： Guifu Yang Chenglin Pu Jingbo Zhang Zhenbang Liu Xiulin Huang Xiaoyu Xu Zhiqiang Ma School of Computer Science and Information Technology Northeast Normal University Key Laboratory of Intelligent Information Processing of Jinlin Universities Changchun Institute of Applied Chemistry Chinese Academy of Sciences

WiFi positioning is the most promising means of indoor positioning in the ***,most WiFi localization techniques can't meet the requirements of the criminal ***,there is a need to make a research on the passive *** paper mainly researched the passive location of WiFi client based on the location *** on the signal attenuation model,fingerprint location algorithm and promiscuous mode,an improved technique of passive positioning is ***,we conducted an in-depth research for the position of the fingerprint algorithm based on RSSI(Received Signal Strength Indication).Then,we propose solutions to optimize the location fingerprint to reduce location time and calculation ***,simulation results and field test data show that symbolic positioning is good and physical positioning can reach high position accuracy.

关键词： WiFi localization location fingerprint identification passive location

来源：评论

学校读者我要写书评

暂无评论

Fine-grained pseudo-code generation method via code feature extraction and transformer

arXiv

引用

arXiv 2021年

作者： Yang, Guang Zhou, Yanlin Chen, Xiang Yu, Chi School of Information Science and Technology Nantong University China Ministry of Industry and Information Technology China The Key Laboratory of Cognitive Computing and Intelligent Information Processing of Fujian Education Institutions Wuyi University China

Pseudo-code written by natural language is helpful for novice developers’ program comprehension. However, writing such pseudo-code is time-consuming and laborious. Motivated by the research advancements of sequence-to-sequence learning and code semantic learning, we propose a novel deep pseudo-code generation method DeepPseudo via code feature extraction and Transformer. In particular, DeepPseudo utilizes a Transformer encoder to perform encoding for source code and then use a code feature extractor to learn the knowledge of local features. Finally, it uses a pseudo-code generator to perform decoding, which can generate the corresponding pseudo-code. We choose two corpora (i.e., Django and SPoC) from real-world large-scale projects as our empirical subjects. We first compare DeepPseudo with seven state-of-the-art baselines from pseudo-code generation and neural machine translation domains in terms of four performance measures. Results show the competitiveness of DeepPseudo. Moreover, we also analyze the rationality of the component settings in DeepPseudo. Copyright © 2021, The Authors. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Study of automated PCNN system based on fruit fly optimization algorithm

引用

Journal of Computational information Systems 2014年第15期10卷 6635-6642页

作者： Zhao, Han Ding, Shifei School of Computer Science and Technology China University of Mining and Technology Xuzhou China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

As the third-generation neural network technology, pulse coupled neural network (PCNN) had used in many fields successfully, but it hindered its popularize that so many parameters of the PCNN need to be set up. This paper presented a new method to automatically determine the parameters of the PCNN based on Fruit Fly Optimization Algorithm (FOA). In proposed method, the information entropy of the image is defined as the fitness function of Fruit Fly Optimization Algorithm when searching the optimal solution. Mutual information is used as the evaluation criteria in the simulation experiments. Contrasted the simulation result with the experiments used the Otsu method, manual adjustment method, genetic algorithm method and PSO method, it shows that the FOA method is better than the others. © 2014 Binary information Press

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Accelerating genetic algorithm for solving graph coloring problem based on CUDA architecture

Communications in Computer and Information Science

引用

Communications in Computer and information Science 2014年 472卷 578-584页

作者： Zhang, Kai Qiu, Ming Li, Lin Liu, Xiaoming School of Computer Science Wuhan University of Science and Technology Wuhan Hubei430081 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System China

ISBN: (纸本)9783662450482

Graph coloring problem (GCP) is a well-known NP-hard combinatorial optimization problem in graph theory. Solution for GCP often finds its applications to various engineering fields. So it is very important to find a feasible solution quickly. Recent years, Compute Unified Device Architecture (CUDA) show tremendous computational power by allowing parallel high performance computing. In this paper, we present a novel parallel genetic algorithm to solve the GCP based on CUDA. The initialization, crossover, mutation and selection operators are designed parallel in threads. Moreover, the performance of our algorithm is compared with the other graph coloring methods using standard DIMACS benchmarking graphs, and the comparison result shows that our algorithm is more competitive with computation time and graph instances size. © Springer-Verlag Berlin Heidelberg 2014.

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

Dog recognition in Public Places Based on Convolutional Neural Network

Dog Recognition in Public Places Based on Convolutional Neur...

引用

2019中国信息产业创新发展大会

作者： Ouyang Jianquan He Hao He Yi Tang Huanrong Xiangtan University College of Information Engineering 411105China Key Laboratory of Intelligent Computing and Information Processing Ministry of Education.College of Information Engineering.Xiangtan UniversityXiangtan 411105China

With the increase in the number of dogs in the city,the scene of the dogs in public places can be seen *** the same time,more and more stray dogs appear in public places where dogs are prohibited,which have a certain impact on the city environment and personal *** view of this,we propose a novel algorithm that combines Dense-SIFT and CNN to solve dog recognition problems in public ***,the image is divided into several grids;then the Dense-SIFT algorithm is used to split and combine the descriptors,and the channel information of the eight directions of the image is extracted as the input of the CNN;finally,we design a CNN based on Adam optimization algorithm and cross-entropy to identify the dog *** Experimental results show that the algorithm can fully combine the advantages of Dense-SIFT and CNN to achieve dog recognition in public places,and the correct rate is 94.2%.

关键词： CNN Dense-SIFT Dog recognition

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：