Authentication email has already been sent, please check your email box: and activate it as soon as possible.
You can login to My Profile and manage your email alerts.
If you haven’t received the email, please:
|
|
There are 233 papers published in subject: > since this site started. |
Select Subject |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
1. Iris segmentation based on active contour model and boundary constraints | |||
YAN Ya-Bing, AN Ling-Ling, KABANO Gilles, WANG Quan | |||
Computer Science and Technology 02 June 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In this paper, we present an iris segmentation method based on active contour model and boundary constraints. First, the proposed method utilizes the localizing region-based active contours model to detect the exterior boundary of the iris region approximately. Then the accurate segmentation together with the boundary constraints could be obtained according to the boundary mask. Following this, the circular Hough transform is adopted to localize the interior boundary of iris region under boundary constraints. Finally, the accurate iris segmentation can be achieved by integrating the interior and exterior boundaries. Extensive experiments demonstrate that the proposed method overcomes the disturbances caused by eyelid and eyelash occlusions, eyeglass and specular reflections, improving the accuracy of iris segmentation effectively. | |||
TO cite this article:YAN Ya-Bing, AN Ling-Ling, KABANO Gilles, et al. Iris segmentation based on active contour model and boundary constraints[OL].[ 2 June 2016] http://en.paper.edu.cn/en_releasepaper/content/4695299 |
2. 3-D Skeleton Recovery via Sparse Representation | |||
WANG Meiyuan,LI Kun,YANG Jingyu,WU Feng,LAI Yu-kun | |||
Computer Science and Technology 17 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Skeleton tracking is a useful and popular application of Kinect. However, it cannot provide accurate reconstructions for complex motions, especially in the case of occlusion. This paper proposes a new 3-D motion recovery method based on low-rank matrix analysis to correct the invalid or corrupt motions. We address this problem by introducing a convex low-rank matrix recovery model, which finds the correct low-rank matrix by fixing its erroneous entries through minimizing the sum of L1-norm and nuclear norm. Experimental results show that our method recovers the corrupted skeleton joints, achieving accurate and smooth reconstructions for complicated motions.????? | |||
TO cite this article:WANG Meiyuan,LI Kun,YANG Jingyu, et al. 3-D Skeleton Recovery via Sparse Representation[OL].[17 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4689753 |
3. Computer supported cooperative work and shared economy: principles & case study | |||
YIN Chuantao,DAVID Bertrand,XIONG Zhang | |||
Computer Science and Technology 17 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In the article we analyze the characteristics of collaborative systems and the characteristics of shared economy supporting systems. Uberisation as present in many applications: Airbnb, Uber, BlablaCar, AMAP, circular economy... needs cooperative system support. We examine this approach from the point of view of ICT and more specifically the HMI (Human Machine Interaction) and CSCW (Computer Supported Cooperative Work (CSCW) and indicate what must be added to collaborative systems to support uberisation. To identify appropriate collaborative model and show how to add new uberisation services to obtain an uberisation supporting platform is also included in the agenda. A case of design of a collaborative application for Carbon Free Parcel Distribution will also be presented and corresponding intermediation algorithms discussed. | |||
TO cite this article:YIN Chuantao,DAVID Bertrand,XIONG Zhang. Computer supported cooperative work and shared economy: principles & case study[OL].[17 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4689362 |
4. Cost-Efficient VM Configuration Algorithm in the Cloud using Mix Scaling Approach | |||
Li Lu,Jiadi Yu | |||
Computer Science and Technology 17 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Benefiting from the pay-per-use pricing model of cloud computing, more and more companies and developers are willing to migrate their services and applications from typical expensive infrastructures to the cloud. However, since there are usually fluctuation in the workload of services or applications, making a cost-efficient VM configurations decision in the cloud remains a critical challenge. Even experienced administrators cannot accurately predict the workload in the future. Instead of scaling out strategy which is usually used, In our paper, we adopt mix scale approach, since the pricing model of cloud provider is convex other than linear that often assumed in past research. Based on this observation, we model a optimization problem aiming to minimize VM renting cost under the constraint of migration delay. Taking advantages of Lyapunov optimization techniques, we propose a mix scale online algorithm which could achieve more cost-efficiency than that of scale out strategy to solve this problem. And our experimental results shows that our mix scale algorithm can save 30.8% and 31.1% cost while controlling migration delay in a reasonable range under low-fluctuation and high-fluctuation workload respectively. | |||
TO cite this article:Li Lu,Jiadi Yu. Cost-Efficient VM Configuration Algorithm in the Cloud using Mix Scaling Approach[OL].[17 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4689395 |
5. Surface Height Map Estimation From a Single Image Using Convolutional Neural Networks | |||
ZHOU Xiaowei, ZHONG Guoqiang, QI Lin, DONG Junyu , MAO Jianzhou | |||
Computer Science and Technology 14 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Surface height map estimation is an important task in high-resolution 3D reconstruction. This task differs from general scene depth estimation in the fact that surface height maps contain more high frequency information or fine details. Existing methods based on radar, laser or other equipments can be used for large-scale scene depth recovery, but might fail in small-scale surface height map estimation. Although some methods are available for surface height reconstruction based on multiple images, e.g. photometric stereo, height map estimation directly from a single image is still a challenging issue. In this paper, we present a novel method based on convolutional neural networks for estimating the height map from a single texture image, without using any other equipments or extra prior knowledge of the image contents. Experimental results based on procedural and real texture datasets show that the proposed algorithm is effective and reliable. | |||
TO cite this article:ZHOU Xiaowei, ZHONG Guoqiang, QI Lin, et al. Surface Height Map Estimation From a Single Image Using Convolutional Neural Networks[OL].[14 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4688898 |
6. Accelerate convolutional neural networks for binary classification: a cascading cost-sensitive feature approach | |||
PANG Jun-Biao, LIN Hui-Huang, DUAN Li-Juan, HUANG Qing-Ming, YIN Bao-Cai | |||
Computer Science and Technology 13 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Convolutional Neural Networks (CNNs) have delivered impressive state-of-the-art performances for many vision tasks, while the computation costs of these networks during test-time are notorious. Empirical results have discovered that CNNs have learned the redundant representations both within and across different layers. When CNNs are applied for binary classification, this article investigate a method to exploit this redundancy across layers, and construct a cascade of classifiers which explicitly balances classification accuracy and hierarchical feature extraction costs.%Rather than reducing redundancy within convolution filters,This method cost-sensitively selects feature points across several layers from trained networks and embeds non-expensive yet discriminative features into a cascade. Experiments on binary classification demonstrate that our framework leads to drastic test-time improvements, e.g., possible $47.2 imes$ speedup for TRECVID upper body detection, $2.82 imes$ speedup for Pascal VOC2007 People detection, $3.72 imes$ for INRIA Person detection % and average $2.5 imes$ speedup for CIFAR-10with less than 0.5% drop in accuracies of the original networks. | |||
TO cite this article:PANG Jun-Biao, LIN Hui-Huang, DUAN Li-Juan, et al. Accelerate convolutional neural networks for binary classification: a cascading cost-sensitive feature approach[OL].[13 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4688871 |
7. Procedural Texture Generation Based on Semantic Descriptions | |||
DONG Jun-Yu, WANG Li-Na, LIU Jun,SUN Xin | |||
Computer Science and Technology 10 May 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Procedural textures are normally generated from mathematical models and have been widely used in computer games and animations for efficient rendering of natural elements, such as wood, marble, stone and other materials. Although the intuitive way to describe procedural texture is to use semantic attributes, there is no connection between procedural models, model parameters and texture semantic descriptions. In this paper, we propose a novel framework for generating procedural textures according to semantic descriptions. First a vocabulary of semantic attributes is collected for describing procedural textures based on extensive psychophysical experiments. Then a multi-label learning method is employed to label more new textures using the semantic attributes. We construct a procedural texture dataset with semantic attributes and further learn a low-dimensional semantic texture space. Finally, for a set of input semantic descriptions, we are able to find a generation model with proper parameters in this space. This model can be used to generate procedural textures that retain the input semantic attributes. Experimental results show that the proposed framework is effective and the generated procedural textures are correlated with the corresponding input semantic descriptions. | |||
TO cite this article:DONG Jun-Yu, WANG Li-Na, LIU Jun, et al. Procedural Texture Generation Based on Semantic Descriptions[OL].[10 May 2016] http://en.paper.edu.cn/en_releasepaper/content/4688895 |
8. An Effective Estimation of Distributed Algorithm for Solving Identical Parallel Machine Scheduling Problem with Precedence Constraints | |||
WU Chuge,WANG Ling,ZHENG Xiaolong | |||
Computer Science and Technology 26 February 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In this paper, an effective estimation of distributed algorithm (eEDA) is proposed to solve the identical parallel machine scheduling problem with precedence constraints (prec-IPMSP). First, the permutation-based encoding scheme is adopted and the earliest finish time (EFT) method is proposed to decode the solution. Second, a new probability model is designed, which describes the relative positions of the jobs. Based on the model, an incremental learning based updating method is developed and a sampling mechanism is proposed to generate feasible solutions with good diversity. In addition, the Taguchi method of design-of-experiment method is used to investigate the effect of key parameters on the performance of the eEDA. Finally, the comparative results of the numerical testing show that the eEDA outperforms the existing algorithm. | |||
TO cite this article:WU Chuge,WANG Ling,ZHENG Xiaolong. An Effective Estimation of Distributed Algorithm for Solving Identical Parallel Machine Scheduling Problem with Precedence Constraints[OL].[26 February 2016] http://en.paper.edu.cn/en_releasepaper/content/4678754 |
9. Study on Circle Detection Algorithm based on Data Dispersion | |||
WEI Youying,SHUAI Liguo,CHEN Huiling | |||
Computer Science and Technology 26 January 2016 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Keywords: circle detection, center coordinates, dispersion To reduce time-consuming, a new algorithm for circle detection is proposed based on data dispersion. The center coordinates and radius can be detected with five steps in this algorithm precisely and quickly; firstly, to reduce the original circle to single-pixel width circle with image processing. Secondly, to calculate the center coordinates with three arbitrary points on the circle. There might be a deviation between the calculated center and real center. Thirdly, to determine a square area for the center coordinates computing with an experimental range and each pixel inside the square is a potential center. Fourthly, to compute the center with distance criterion and the center coordinate is determined when the variance reaches the minimum. Lastly, the radius is equal to the means of the distance vector with the minimum variance. Experiments are conducted and the results show that, comparing with the traditional Hough transform, the new algorit????? | |||
TO cite this article:WEI Youying,SHUAI Liguo,CHEN Huiling. Study on Circle Detection Algorithm based on Data Dispersion[OL].[26 January 2016] http://en.paper.edu.cn/en_releasepaper/content/4677465 |
10. Evaluating user's influence based on temporal topic and connection state in Online Social Networks | |||
Wang Feng,Jiang Wenjun,Wang Guojun,Meng Dacheng | |||
Computer Science and Technology 16 December 2015 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Social influence means that the user'sactions can induce his/her friend to behave in a similar way, this phenomenon is essential for various applications in online social networks (OSNs). Messages received by a user mainly depend on whom the user follows, so each followed use can be treated as information sources. In order to receive messages more efficient, user should to know how to find high social influence user. To address this problem, this paper proposes a new evaluation model of user's social influence. We propose the temporal-topic influence evaluation model to incorporate topic distribution information andtime factor.By analyzing the temporal topic interest distribution, we can evaluate user's social influence more accurately and efficiently. And our model also considers the connection state information determines user's social influence. The evaluation results showuser's influence I(u)at timet_i. We verify the efficiency and accuracy of our model by experiments, and the experiment results show that the proposed approach can efficiently infer user's social influence. | |||
TO cite this article:Wang Feng,Jiang Wenjun,Wang Guojun, et al. Evaluating user's influence based on temporal topic and connection state in Online Social Networks[OL].[16 December 2015] http://en.paper.edu.cn/en_releasepaper/content/4665863 |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
|
About Sciencepaper Online | Privacy Policy | Terms & Conditions | Contact Us
© 2003-2012 Sciencepaper Online. unless otherwise stated