检索结果-南通市图书馆

A Gaussian copula regression model for movie box-office revenues prediction

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2017年第9期60卷 36-49页

作者： Junwen DUAN Xiao DING Ting LIU Research Center for Social computing and Information Retrieval Harbin Institute of Technology

In this article, we revisit the task of movie box-office revenues prediction using multi-type *** movie box-office revenues are affected by numerous factors. Previous work with discriminative models assumes these factors are identically and independently distributed. The correlations between these factors are rarely considered, which limited the performances of discriminative models in this task. To address these problems, we investigate a novel Gaussian copula regression model. Based on this model, we do not need to make any prior assumptions about the marginal distributions of the features. In particular, we perform a cumulative probability estimation on each of the smoothed features. The estimation learns the marginal distributions and maps all features into a uniform vector space. Sequentially, we bridge the marginal distributions with a copula function to create their joint distribution, and learn the dependency structure between them. Moreover, we propose a computational-efficient approximate algorithm for responsible variable inference. Experimental results on two movie datasets from Chinese and U.S. market show that our approach outperforms strong discriminative regression baselines.

关键词： Gaussian copula movie box-office revenue multi-variate regression text regression social media

An Innovative Methodology of Product Design from Nature

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Bionic Engineering 2008年第1期5卷 75-84页

作者： Hui-I Wen Shu-jun Zhang Kevin Hapeshi Xiao-feng Wang Department of Visual Communication Design Chaoyang University of Technology Taichung County 41394 Taiwan Department of Multi-media and computing Gloucestershire University Cheltenham GL50 2HR UK Warwick Manufacturing Group School of Engineering The University of Warwick Coventry CV4 7AL UK

Nature is an information sourcebook of behaviour, function, colour and shape which can inspire visual design and invention. Studying the form and functional characteristics of a natural object can provide inspiration for product design and help to improve the marketability of manufactured products. The inspiration can be triggered either by direct observation or captured by three-dimensional （3D） digitising techniques to obtain superficial information （geometry and colour）. An art designer often creates a concept in the form of a two-dimensional （2D） sketch while engineering methods lead to a point cloud in 3D. Each has its limitations in that the art designer commonly lacks the knowledge to build a final product from a 2D sketch and the engi- neering designer＇s 3D point clouds may not be very beautiful. We propose a method for Product Design from Nature （PDN）, coupling aesthetic intent and geometrical characteristics, exploring the interactions between designers and nature＇s systems in PDN. We believe that this approach would considerably reduce the lead time and cost of product design from nature.

关键词： bio-inspiration design conceptual design designer sketch inspiration forms aesthetic design from nature

EFECL:Feature encoding enhancement with contrastive learning for indoor 3D object detection

在线全文

学校读者我要写书评

暂无评论

Computational Visual media 2023年第4期9卷 875-892页

作者： Yao Duan Renjiao Yi Yuanming Gao Kai Xu Chenyang Zhu School of computing National University of Defense TechnologyChangsha 410000China

Good proposal initials are critical for 3D object detection ***,due to the significant geometry variation of indoor scenes,incomplete and noisy proposals are inevitable in most *** feature information among these“bad”proposals may mislead the *** learning provides a feasible way for representing proposals,which can align complete and incomplete/noisy proposals in feature *** aligned feature space can help us build robust 3D representation even if bad proposals are ***,we devise a new contrast learning framework for indoor 3D object detection,called EFECL,that learns robust 3D representations by contrastive learning of proposals on two different ***,we optimize both instance-level and category-level contrasts to align features by capturing instance-specific characteristics and semantic-aware common ***,we propose an enhanced feature aggregation module to extract more general and informative features for contrastive *** on ScanNet V2 and SUN RGB-D benchmarks demonstrate the generalizability and effectiveness of our method,and our method can achieve 12.3%and 7.3%improvements on both datasets over the benchmark *** code and models are publicly available at https://***/YaraDuan/EFECL.

关键词： indoor scene object detection contrastive learning feature enhancement

Pricing Based Power Control for Inter-Cell UAV-Assisted Vehicle-to-Vehicle Underlay Communication

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2019年第1期16卷 57-68页

作者： Qin Wang Ye Chen Shu Yin Lei Tian Yongan Guo Jiangsu Key laboratory of Wireless Communications Nanjing University of Posts and Telecommunications School of Engineering & computing Sciences New York Institute of Technology College of Overseas Education Nanjing University of Posts and Telecommunications

The high-capacity vehicle-to-vehicle(V2 V) communication provides a promising solution to support ubiquitous media streaming and content sharing among *** extend the V2 V links to multiple cells and manage the inter-cell interference,we proposed an UAV-assisted inter-cell V2 V communication model,in which a shared UAV node is placed in the center of V2 V *** charging the V2 V users underlay spectrum access fee,the cellular network earn profit at the cost of encountering co-channel interference from V2 V links.A Stackelberg game is formulated to model the interactions between the V2 V links and the cellular links,which are the game follower and the leader *** utility functions are maximized in terms of accessing price as well as transmit power of V2 V users and UAV *** evaluations verify that the power-price tradeoff between V2 V network and cellular network has significant potentials to enhance their utility.

关键词： UAV-assisted communication inter-cell V2V Stackelberg game IoT economics

Unsupervised image translation with distributional semantics awareness

在线全文

学校读者我要写书评

暂无评论

Computational Visual media 2023年第3期9卷 619-631页

作者： Zhexi Peng He Wang Yanlin Weng Yin Yang Tianjia Shao State Key Lab of CAD&CG Zhejiang UniversityHangzhou 310058China School of computing University of LeedsLeedsUK School of computing Clemson UniversityClemsonUSA

Unsupervised image translation(UIT)studies the mapping between two image *** such mappings are under-constrained,existing research has pursued various desirable properties such as distributional matching or two-way *** this paper,we re-examine UIT from a new perspective:distributional semantics consistency,based on the observation that data variations contain semantics,e.g.,shoes varying in ***,the semantics can be multi-dimensional,e.g.,shoes also varying in style,functionality,*** two image domains,matching these semantic dimensions during UIT will produce mappings with explicable correspondences,which has not been investigated *** propose distributional semantics mapping(DSM),the first UIT method which explicitly matches semantics between two *** show that distributional semantics has been rarely considered within and beyond UIT,even though it is a common problem in deep *** evaluate DSM on several benchmark datasets,demonstrating its general ability to capture distributional *** comparisons show that DSM not only produces explicable mappings,but also improves image quality in general.

关键词： generative adversarial networks(GANs) manifold alignment unsupervised learning image-to-image translation distributional semantics

ChatGPT Chats on Computational Experiments:From Interactive Intelligence to Imaginative Intelligence for Design of Artificial Societies and Optimization of Foundational Models

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2023年第6期10卷 1357-1360页

作者： Xiao Xue Xiangning Yu Fei-Yue Wang College of Intelligence and computing Tianjin UniversityTianjin 300350China State Key laboratory for Management and Control of Complex Systems the Institute of Automdtion of Chinese Academg of SciencesBeijing 100190China

POWERED by the rapid development of Internet,the penetration of the Internet of Things,the emergence of big data,and the rise of social media,more and more complex systems are exhibiting the characteristics of social,physical,and information *** systems are known as cyber-physicalsocial systems(CPSS)[1],[2].These CPSS face unprecedented challenges in design,analysis,management,control and integration due to their involvement with human and social factors[3],[4].To cope with this challenge,there are two main approaches to CPSS research.

关键词： Intelligence Foundation involvement

A Robust Image Watermarking Scheme Using Z-Transform, Discrete Wavelet Transform and Bidiagonal Singular Value Decomposition

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2019年第1期58卷 263-285页

作者： N.Jayashree R.S.Bhuvaneswaran Ramanujan computing Centre Anna UniversityChennai600025India

Watermarking is a widely used solution to the problems of authentication and copyright protection of digital media especially for images,videos,and audio *** is one of the emerging techniques adopted in image watermarking schemes due to its intrinsic cryptographic *** paper proposes a new chaotic hybrid watermarking method combining Discrete Wavelet Transform(DWT),Z-transform(ZT)and Bidiagonal Singular Value Decomposition(BSVD).The original image is decomposed into 3-level DWT,and then,ZT is applied on the HH3 and HL3 *** watermark image is encrypted using Arnold Cat *** for the watermark and transformed original image were computed,and the watermark was embedded by modifying singular values of the host image with the singular values of the watermark *** of the proposed scheme was examined using standard test images and assessed against common signal processing and geometric *** indicated that the proposed method is transparent and highly robust.

关键词： Digital watermarking chaotic mapping z-transform arnold cat map discrete wavelet transform(DWT) bidiagonal singular value decomposition(BSVD)

Quality measures of reconstruction filters for stereoscopic volume rendering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual media 2016年第1期2卷 19-30页

作者： David A.T.Roberts Ioannis Ivrissimtzis School of Engineering and computing Sciences Durham University

In direct volume rendering(DVR), the choice of reconstruction filter can have a significant effect on the visual appearance of the images produced and thus, on the perceived quality of a DVR rendered scene. This paper presents the results of a subjective experiment where participants stereoscopically viewed DVR rendered scenes and rated their subjective *** statistical analysis of the results focuses on the relationship between the quality of the stereoscopic scene and properties of the filters such as post-aliasing and smoothing, as well as the relationship between the quality of the stereoscopic scene and properties of the rendered images such as shape *** experiment evaluated five reconstruction filters on four different volumetric datasets. Participants rated the stereoscopic scenes on four quality measures:depth quality, depth layout, lack of jaggyness, and sharpness. The results show that the correlation between the quality measures and post-aliasing and smoothing, which are properties associated with each reconstruction filter, is moderate and statistically insignificant. On the other hand, the correlation between the quality measures and compactness, which is a property specific to each rendered image, is strong and statistically significant.

关键词： direct volume rendering(DVR) reconstruction filters stereoscopic user evaluations

在线全文

学校读者我要写书评

暂无评论

Inflatable Models

Journal of Computer Science & Technology 2006年第2期21卷 154-158页

作者：李凌 Vasily Volkov Department of computing Curtin University of Technology Perth WA 6845 Australia

A physically-based model is presented for the simulation of a new type of deformable objects-inflatable objects, such as shaped balloons, which consist of pressurized air enclosed by an elastic surface. These objects have properties inherent in both 3D and 2D elastic bodies, as they demonstrate the behaviour of 3D shapes using 2D formulations. As there is no internal structure in them, their behaviour is substantially different from the behaviour of deformable solid objects. We use one of the few available models for deformable surfaces, and enhance it to include the forces of internal and external pressure. These pressure forces may also incorporate buoyancy forces, to allow objects filled with a low density gas to float in denser media. The obtained models demonstrate rich dynamic behaviour, such as bouncing, floating, deflation and inflation.

关键词： inflatable models physically based modelling physically based animation