检索结果-内蒙古大学图书馆

Coarse-to-Fine Feedback Guidance Based Stereo Image Quality Assessment Considering Dominant Eye Fusion

ieee TRANSACTIONS ON MULTIMEDIA 2023年 25卷 8855-8867页

作者： Chang, Yongli Li, Sumei Liu, Anqi Jin, Jie Xiang, Wei Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China La Trobe Univ Sch Engn & Math Sci Melbourne Vic 3086 Australia

Considering that the human brain always follows a coarse-to-fine (low-to-high spatial frequency) visual processing and fusion mechanism, we propose a coarse-to-fine feedback guidance based stereo image quality assessment (SIQA) network which considers a coarse-to-fine feedback guidance and adaptive dominant eye mechanism. The proposed network consists of two main sub-network streams, each of which has three branches to extract low, middle and high spatial frequency information in parallel. To better realize the guidance of the high-level features in the low spatial frequency branch to the low-level features in the high spatial frequency branch, an information feedback guidance module (IFGM) is proposed, which realizes a top-down guidance mechanism in each sub-network stream. Simultaneously, according to the theory of ocular dominance in human visual system (HVS), we design an adaptive bi-directional parallax-based binocular fusion module (BPBFM), which synthesizes two types of fusion feature by taking the left and right view features as dominant eye input. Furthermore, in order to obtain the better perceptual quality of stereo images, we design a weighted fusion strategy to weigh the quality scores from the two types of fusion features obtained by using an ensemble model with two multi-layer perceptrons (MLPs). The experimental results on four public stereo image datasets show that the proposed method is superior to the mainstream metrics and achieves an excellent performance.

关键词： visualization Feature extraction Image quality Adaptive systems Streaming media data mining Bidirectional control Stereo image quality assessment coarse-to-fine feedback guidance bi-directional parallax attention mechanism convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Fine-Grained Feature Generation for Generalized Zero-Shot Video Classification

引用

ieee TRANSACTIONS ON IMAGE PROCESSING 2023年 32卷 1599-1612页

作者： Hong, Mingyao Zhang, Xinfeng Li, Guorong Huang, Qingming Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing 100049 Peoples R China

Generalized zero-shot video classification aims to train a classifier to classify videos including both seen and unseen classes. Since the unseen videos have no visual information during training, most existing methods rely on the generative adversarial networks to synthesize visual features for unseen classes through the class embedding of category names. However, most category names only describe the content of the video, ignoring other relational information. As a rich information carrier, videos include actions, performers, environments, etc., and the semantic description of the videos also express the events from different levels of actions. In order to use fully explore the video information, we propose a fine-grained feature generation model based on video category name and its corresponding description texts for generalized zero-shot video classification. To obtain comprehensive information, we first extract content information from coarse-grained semantic information (category names) and motion information from fine-grained semantic information (description texts) as the base for feature synthesis. Then, we subdivide motion into hierarchical constraints on the fine-grained correlation between event and action from the feature level. In addition, we propose a loss that can avoid the imbalance of positive and negative examples to constrain the consistency of features at each level. In order to prove the validity of our proposed framework, we perform extensive quantitative and qualitative evaluations on two challenging datasets: UCF101 and HMDB51, and obtain a positive gain for the task of generalized zero-shot video classification.

关键词： visualization Semantics Task analysis Training Generative adversarial networks Feature extraction data models Zero-shot learning feature generation video classification

来源：评论

学校读者我要写书评

暂无评论

Analysis of Behavioural data of Customer for the E-Commerce Platform by using Machine Learning Approach

Analysis of Behavioural Data of Customer for the E-Commerce ...

引用

2023 International Conference on Computational Intelligence and Sustainable Engineering Solution, CISES 2023

作者： Maurya, Ayush Pratap, Saurabh Pratap, Prabal Dwivedi, Ashish Shambhunath Institute of Engineering and Technology Department of Computer Science and Engineering UP Prayagraj India Department of Mechanical Engineering U.P Varanasi India Defence Research and Development Organisation U.P Kanpur India Jindal Global Business School Haryana Sonepat131001 India

ISBN: (纸本)9798350323917

This research analyses the effect of numerous factors that allow E-commerce platforms to obtain previous knowledge about customer purchase tendencies. This research improves customer behaviour, product discoverability, and warehouse maintenance. Demanding product categories assist maintain the warehouse and generate sales ideas by offering an appropriate discount to attract consumers and promote same-day delivery. The system uses an e-commerce company database to store consumer purchase information. Our model analyses the data to classify customers and commodities. Our study includes descriptive, predictive, and prescriptive analytics to anticipate e-commerce sales. The descriptive study includes data cleaning, preprocessing, and visualization to analyze gender, city tier, spending money, age, city, marital status, financial position, brand cost, and product categories these are the attributes of dataset. To advance to predictive analytics, machine learning techniques such as naive Bayes classification, support vector classification, logistical regression, decision tree, KNN classification, and random forest classification are applied to the dataset to anticipate product sales. This proposed research reduces e-waste and promotes sustainable development by anticipating product sales. © 2023 ieee.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Dual Relation Network for Scene Text Recognition

引用

ieee TRANSACTIONS ON MULTIMEDIA 2023年 25卷 4094-4107页

作者： Li, Ming Fu, Bin Chen, Han He, Junjun Qiao, Yu Chinese Acad Sci Shenzhen Inst Adv Technol Guangdong Hong Kong Macao Joint Lab Human Machine Hong Kong 518055 Guangdong Peoples R China Shanghai AI Lab Shanghai 518055 Peoples R China

Local visual and long-range contextual features yield two complementary cues for human reading text in natural scene. Existing scene text recognition methods mainly extract local features at a low level and then model long-range dependencies at a high level, this sequential pipeline may be sub-optimal to construct complete and effective representation. Except for high-level features, long-range contextual relation is of importance in low-level features as well since it can help separate different characters based on the intervals between characters and thus enhance the character features. To address this issue, we develop a dual relation module to extract complementary features in a parallel manner for scene text recognition, which consists of a local visual branch and a long-range contextual branch. The local visual branch employs a topological-aware operation to model intra-character characteristic and extract discriminative features of different characters. Meanwhile, the long-range contextual branch utilizes a simple but effective strategy to incorporate inter-character relations into feature maps. Our dual relation module is a plug-and-play block which can be easily incorporated into modern deep architectures. Experimental results demonstrate that our methods achieved top performance on several standard benchmarks. Code and models will become publicly available in the future.

关键词： Feature extraction Text recognition visualization Task analysis Context modeling data mining Convolution Scene text recognition scene optical character recognition deep learning

来源：评论

学校读者我要写书评

暂无评论

Immersive Viewing of Dento-facial Medical data in Virtual Reality 14

Immersive Viewing of Dento-facial Medical Data in Virtual Re...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Bhat, Shreenidhi H. Hareesh, K.S. Kamath, Abhay T. Kudva, Adarsh Manipal Institute of Technology Manipal Academy of Higher Education Dept. Data Science and Computer Applications Manipal India Consultant-Oral and Maxillofacial Surgery Kmc Hospitals Mangaluru India Manipal College of Dental Sciences Manipal Academy of Higher Education Dept. Oral and Maxillofacial Surgery Manipal India

ISBN: (纸本)9798350335095

Virtual reality (VR) requires dynamic and responsive rendering technologies to offer an immersive user experience. Failing, users may experience motion sickness due to motion lag in virtual space. This work aims to create a framework with a higher framerate to decrease the motion sickness in turn increasing the virtual world viewing quality. Essentially, an interactive environment is developed between existing volume rendering engines OpenGL and GPU-based ray casting techniques for rendering the dentofacial data volume. The developed visualization methods are incorporated into Head Mount Device(HMD) to offer immersive effects. The performance of the VR rendering application measured in terms of framerate was observed to be 125 frames per second, significantly higher than the current HMDs framerate. In addition, the adaptation of rendered image quality is activated by adjusting the sampling distance and jitter to complete the rendering process in the allotted time. The achieved result is expected to offer satisfactory comfort to the viewer with reduced motion sickness, particularly in pre-surgical planning for dentofacial surgery. © 2023 ieee.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

360MVSNet: Deep Multi-view Stereo Network with 360° Images for Indoor Scene Reconstruction 23

360MVSNet: Deep Multi-view Stereo Network with 360° Images ...

引用

23rd ieee/CVF Winter Conference on Applications of Computer Vision (WACV)

作者： Chiu, Ching-Ya Wu, Yu-Ting Shen, I-Chao Chuang, Yung-Yu Natl Taiwan Univ New Taipei Taiwan Natl Taipei Univ New Taipei Taiwan Univ Tokyo Tokyo Japan

ISBN: (纸本)9781665493468

Recent multi-view stereo methods have achieved promising results with the advancement of deep learning techniques. Despite of the progress, due to the limited fields of view of regular images, reconstructing large indoor environments still requires collecting many images with sufficient visual overlap, which is quite labor-intensive. 360 degrees images cover a much larger field of view than regular images and would facilitate the capture process. In this paper, we present 360MVSNet, the first deep learning network for multi-view stereo with 360 degrees images. Our method combines uncertainty estimation with a spherical sweeping module for 360 degrees images captured from multiple viewpoints in order to construct multi-scale cost volumes. By regressing volumes in a coarse-to-fine manner, high-resolution depth maps can be obtained. Furthermore, we have constructed EQMVS, a large-scale synthetic dataset that consists of over 50K pairs of RGB and depth maps in equirectangular projection. Experimental results demonstrate that our method can reconstruct large synthetic and real-world indoor scenes with significantly better completeness than previous traditional and learning-based methods while saving both time and effort in the data acquisition process.

关键词： Deep learning Learning systems visualization Uncertainty Costs data acquisition Estimation

来源：评论

学校读者我要写书评

暂无评论

UltraStar:A Lightweight Simulator of Ultra-Dense LEO Satellite Constellation Networking for 6G

引用

ieee/CAA Journal of Automatica Sinica 2023年第3期10卷 632-645页

作者： Xiaoyu Liu Ting Ma Zhixuan Tang Xiaohan Qin Haibo Zhou Xuemin(Sherman)Shen School of Electronic Science and Engineering Nanjing UniversityNanjing 210023China Department of Electrical and Computer Engineering University of WaterlooWaterloo N2L 3G1Canada IEEE

The mega-constellation network has gained significant attention recently due to its great potential in providing ubiquitous and high-capacity connectivity in sixth-generation(6G)wireless communication ***,the high dynamics of network topology and large scale of mega-constellation pose new challenges to the constellation simulation and performance *** this paper,we introduce UltraStar,a lightweight network simulator,which aims to facilitate the complicated simulation for the emerging mega-constellation of unprecedented ***,a systematic and extensible architecture is proposed,where the joint requirement for network simulation,quantitative evaluation,data statistics and visualization is fully *** characterizing the network,we make lightweight abstractions of physical entities and models,which contain basic representatives of networking nodes,structures and protocol ***,to consider the high dynamics of Walker constellations,we give a two-stage topology maintenance method for constellation initialization and orbit ***,based on the discrete event simulation(DES)theory,a new set of discrete events is specifically designed for basic network processes,so as to maintain network state changes over ***,taking the first-generation Starlink of 11927 low earth orbit(LEO)satellites as an example,we use UltraStar to fully evaluate its network performance for different deployment stages,such as characteristics of constellation topology,performance of end-to-end service and effects of network-wide traffic *** simulation results not only demonstrate its superior performance,but also verify the effectiveness of UltraStar.

关键词： Discrete event simulation(DES) mega-constellation network dynamics performance evaluation simulation architecture design

来源：评论

学校读者我要写书评

暂无评论

Study on Forest Fire Diffusion Method Based on Cellular Automata 30

Study on Forest Fire Diffusion Method Based on Cellular Auto...

引用

30th International Conference on Geoinformatics, Geoinformatics 2023

作者： Zhang, Tonghe Meng, Qingxiang Liu, Jinchun Liang, Qifan Wuhan University School of Remote Sensing and Information Engineering Wuhan China Wuhan University School of Geodesy and Geomatics Wuhan China Wuhan University School of Computer Science Wuhan China

ISBN: (纸本)9798350344240

As one of the serious natural disasters, forest fire has the characteristics of strong suddenness, hard to detect, wide range of disaster, difficult to extinguish and heavy losses. Frequent forest fires put forward an urgent need for dynamic monitoring of forest fire diffusion and prediction of forest fire behavior. This paper constructs a forest fire diffusion model based on cellular automata, and outputs the visualization results of forest fire diffusion by inputting environmental factors such as terrain data, meteorological data and vegetation type data and the location of the fire point. Finally, the model is used to simulate the forest fire diffusion process in Xichang City, Sichuan Province, and the validity of the model is verified by remote sensing images. The results show that the over-fire area obtained by the simulation experiment is similar to the spatial distribution of the over-fire area extracted by Sentinel-2 satellite, which proves that the cellular automata model constructed in this paper has high accuracy. © 2023 ieee.

关键词： Cellular automata

来源：评论

学校读者我要写书评

暂无评论

AN AUTOMATED PIPELINE TO CREATE AN ATLAS OF IN SITU HYBRIDIZATION GENE EXPRESSION data IN THE ADULT MARMOSET BRAIN 20

AN AUTOMATED PIPELINE TO CREATE AN ATLAS OF IN SITU HYBRIDIZ...

引用

20th ieee International Symposium on Biomedical Imaging (ISBI)

作者： Poon, Charissa Rachmadi, Muhammad Febrian Byra, Michal Schlachter, Matthias Xu, Binbin Shimogori, Tomomi Skibbe, Henrik RIKEN Ctr Brain Sci Brain Image Anal Unit Wako Saitama Japan Polish Acad Sci Inst Fundamental Technol Res Warsaw Poland Univ Montpellier IMT Mines Ales EuroMov Digital Hlth Mot Ales France RIKEN Ctr Brain Sci Lab Mol Mech Brain Dev Wako Saitama Japan

ISBN: (纸本)9781665473583

We present the first automated pipeline to create an atlas of in situ hybridization gene expression in the adult marmoset brain in the same stereotaxic space. The pipeline consists of segmentation of gene expression from microscopy images and registration of images to a standard space. Automation of this pipeline is necessary to analyze the large volume of data in the genome-wide whole-brain dataset, and to process images that have varying intensity profiles and expression patterns with minimal human bias. To reduce the number of labelled images required for training, we develop a semi-supervised segmentation model. We further develop an iterative algorithm to register images to a standard space, enabling comparative analysis between genes and concurrent visualization with other datasets, thereby facilitating a more holistic understanding of primate brain structure and function.

关键词： contrastive learning gene atlas segmentation semi-supervised learning registration

来源：评论

学校读者我要写书评

暂无评论

STMotionFormer: Language Guided 3D Human Motion Synthesis with Spatial Temporal Transformer

STMotionFormer: Language Guided 3D Human Motion Synthesis wi...

引用

2023 ieee International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Lv, Yifan Du, Xiaoqin Zhou, Jiashuang Liu, Yongqi Xu, Han School of Computer Science and Artificial Intelligence Wuhan Textile University Wuhan China School of Computer Science and Artificial Intelligence Wuhan Textile University Hubei Provincial Engineering Research Center for Intelligent Textile and Fashion Wuhan China School of Journalism and Information Communication Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798350337020

Inspired by the Transformer's adaptive capability in natural language processing and computer vision domains, and supported by CLIP's strong semantic prior knowledge, we propose a novel generative model named STMotionFormer based on improved Transformer block and CLIP for language-guided 3D skeleton-based human motion synthesis. The skeleton sequence, as a kind of spatial temporal dynamic data, not only contains temporal information but also consists of a natural graph structure by different body parts. We focus on this, propose Temporal Transformer block (T-Former) applies attention mechanism along the temporal dimension to capture the long-term relationship, and propose spatial Transformer block (S-Former) applies attention mechanism in the spatial dimension to aggregate and update joints' spatial features. Therefore, our approach can automatically explore both spatial and temporal information, which can generate skeleton sequence with reasonable graph structure as well as global time relationship. Moreover, powerful semantic prior knowledge of CLIP are injected into our motion-language joint manifold. We evaluate our method on the public KIT motion language dataset that contains manually annotated 3D pose sequences. Experimental results show that our model outperforms the state-of-the-art in terms of APE and AVE respectively. Qualitative visualization results indicate that our model can generate human motions that are more compatible with semantic information even than the GroundTruth. © 2023 ieee.

关键词： CLIP Language Guided Motion synthesis Spatial Temporal Transformer

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：