检索结果-内蒙古大学图书馆

Lighting and Shadow Techniques for Realistic 3D Synthetic Object Compositing in images 2018

Lighting and Shadow Techniques for Realistic 3D Synthetic Ob...

2nd International conference on Digital Signal processing (ICDSP) / International conference on Computer Graphics and Virtuality (ICCGV)

作者： Mihut, Ana Davison, Richard Ushaw, Gary Morgan, Graham Newcastle Univ Sch Comp Urban Sci Bldg Newcastle Upon Tyne Tyne & Wear England

ISBN: (纸本)9781450364027

Artificial 3D object compositing has recently gained interest in the context of emerging augmented and virtual reality technology. The problem takes as input the initial scene and places a 3D object anywhere within it, rendering a realistic output. This approach is challenging in the absence of prior knowledge of scene geometry, or user annotation to approximate it. Single images of a scene may not reveal all the present illumination conditions, nor reliable depth information. The state-of-the-art methods do not account for compositing 3D objects in between existing scene topology, nor cater for the inclusion of objects behind transparent areas within the original scene. We contribute a framework capable of realistic 3D object compositing with physically accurate soft shadows into a 2D scene derived from a stereo image pair which does not rely on the user's scene annotations for recreating a realistic result or on the availability of RGB-D capable hardware. The benefit of this approach is that hardware information and user knowledge are not required factors in either the 3D reconstruction stage, nor the compositing stage of the framework. We conclude with a stepby-step analysis across all of the stages of the pipeline.

关键词： Inverse rendering stereo mapping statistical shadow mapping

来源：评论

学校读者我要写书评

暂无评论

Generating High Resolution Climate Change Projections through Single image Super-Resolution: An Abridged Version 27

Generating High Resolution Climate Change Projections throug...

引用

27th International Joint conference on Artificial Intelligence (IJCAI)

作者： Vandal, Thomas Kodra, Evan Ganguly, Sangram Michaelis, Andrew Nemani, Ramakrishna Ganguly, Auroop R. Northeastern Univ Civil & Environm Engn Boston MA 02115 USA risQ Inc Boston MA USA NASA Ames Res Ctr Washington DC USA Bay Area Environm Res Inst Moffett Field CA USA Univ Corp Monterey CA USA NASA Adv Supercomp Div Washington DC 20546 USA

ISBN: (纸本)9780999241127

The impacts of climate change are felt by most critical systems, such as infrastructure, ecological systems, and power-plants. However, contemporary Earth System Models (ESM) are run at spatial resolutions too coarse for assessing effects this localized. Local scale projections can be obtained using statistical downscaling, a technique which uses historical climate observations to learn a low-resolution to high-resolution mapping. The spatio-temporal nature of the climate system motivates the adaptation of super-resolution image processing techniques to statistical downscaling. In our work, we present DeepSD, a generalized stacked super resolution convolutional neural network (SRCNN) framework with multi-scale input channels for statistical downscaling of climate variables. A comparison of DeepSD to four state-of-the-art methods downscaling daily precipitation from 1 degree (100km) to 1/8 degrees (12.5km) over the Continental United States. Furthermore, a framework using the NASA Earth Exchange (NEX) platform is discussed for downscaling more than 20 ESM models with multiple emission scenarios.

关键词： Climate change

来源：评论

学校读者我要写书评

暂无评论

Action Recognition Using the 3D Dense Micro-Block Difference 2

Action Recognition Using the 3D Dense Micro-Block Difference

引用

conference on Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies II

作者： Voronin, V. Pismenskova, M. Zelensky, A. Cen, Yigang Nadykto, A. Egiazarian, K. Don State Tech Univ Lab Math Methods Image Proc & Intelligent Comp Vi Rostov Na Donu Russia Moscow State Univ Technol STANKIN Moscow Russia Beijing Jiaotong Univ Inst Informat Sci Beijing Peoples R China Tampere Univ Technol Dept Signal Proc Tampere Finland

ISBN: (数字)9781510621886

ISBN: (纸本)9781510621886

This paper describes a framework for action recognition which aims to recognize the goals and activities of one or more human from a series of observations. We propose an approach for the human action recognition based on the 3D dense micro-block difference. The proposed algorithm is a two-stage procedure: (a) image preprocessing using a 3D Gabor filter and (b) a descriptor calculation using 3D dense micro-block difference with SVM classifier. At the first step, an efficient spatial computational scheme designed for the convolution with a bank of 3D Gabor filters is present. This filter intensifies motion using a convolution for a set of 3D patches and arbitrarily-oriented anisotropic Gaussian. For preprocessed frames, we calculate the local features such as 3D dense micro-block difference (3D DMD), which capture the local structure from the image patches at high scales. This approach is processing the small 3D blocks with different scales from frames which capture the microstructure from it. The proposed image representation is combined with fisher vector method and linear SVM classifier. We evaluate the proposed approach on the UCF50, HMDB51 and UCF101 databases. Experimental results demonstrate the effectiveness of the proposed approach on video with a stochastic textures background with comparisons of the state-of-the-art methods.

关键词： action recognition texture micro-block difference 3D Gabor filter

来源：评论

学校读者我要写书评

暂无评论

A Data-Driven Approach for 3D Human Body Pose Reconstruction from a Kinect Sensor 2

A Data-Driven Approach for 3D Human Body Pose Reconstruction...

引用

2018 2nd International conference on Computer Graphics and Digital image processing, CGDIP 2018

作者： Yao, Li Peng, Xiaodong Guo, Yining Ni, Hongjie Wan, Yan Yan, Cairong School of Computer Science and Technology Donghua University No.2999 Renmin North Road Songjiang District Shanghai China

In the study of virtual fitting techniques, human body modeling has always occupied a very important position. Whether a model that is roughly consistent with users can be established has a direct impact on the fitting experience of users. Based on this, we propose an automatic human body pose registration algorithm which can efficiently construct a posed model using priori data and a single Kinect sensor, and provide a good foundation for the later shape registration. Because of the complexity of the human body, there are so many methods that just reconstruct a human body model but no rigged animation skeleton inside. To solve this, we use SMPL which is a recently published statistical body model to fit 3D joints acquired by a Kinect sensor. And finally we project the posed model to the corresponding person in the color image to improve the fitting experience. Our experiments show that the speed and the estimation error of the algorithm are within the tolerance of virtual fitting. © Published under licence by IOP Publishing Ltd.

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

Educational video classification by using a transcript to image transform and supervised learning

Educational video classification by using a transcript to im...

引用

7th International conference on image processing Theory, Tools and Applications, IPTA 2017

作者： Chatbri, Houssem Oliveira, Marlon McGuinness, Kevin Little, Suzanne Kameyama, Keisuke Kwan, Paul Sutherland, Alistair O'Connor, Noel E. Insight Centre for Data Analytics Dublin City University Ireland School of Computing Dublin City University Ireland Faculty of Engineering Information and Systems University of Tsukuba Japan School of Science and Technology University of New England NSW Australia

ISBN: (纸本)9781538618417

In this work, we present a method for automatic topic classification of educational videos using a speech transcript transform. Our method works as follows: First, speech recognition is used to generate video transcripts. Then, the transcripts are converted into images using a statistical cooccurrence transformation that we designed. Finally, a classifier is used to produce video category labels for a transcript image input. For our classifiers, we report results using a convolutional neural network (CNN) and a principal component analysis (PCA) model. In order to evaluate our method, we used the Khan Academy on a Stick dataset that contains 2,545 videos, where each video is labeled with one or two of 13 categories. Experiments show that our method is effective and strongly competitive against other supervised learning-based methods. © 2017 IEEE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

An adaptive threshold for change detection methods using a windowed entropy-based criterion - Application to fault-tolerant fusion in collaborative mobile robotics

An adaptive threshold for change detection methods using a w...

引用

conference on Control and Fault-Tolerant Systems (SysTol)

作者： Bilal Daass Denis Pomorski Kamel Haddadi Univ. Lille Signal et Automatique de Lille Lille France Univ. Lille UMR 8520- IEMN Lille France

ISBN: (纸本)9781728103815

Nowadays, process supervision occupies an important place in quality control, cooperative localization in mobile robotics, video and image processing, and intelligent system design, to name a few. Indeed, any failure of such processes can reduce performance and have serious consequences. The development of statistical methods, capable of detecting and locating anomalies in these dynamic systems as quickly as possible, is of real interest. In this context, we have proposed in a previous study a reformulation of the change detection strategy using an entropy-based criterion. Our approach allowed the calculation of an adaptive threshold, unlike the Bayes criterion. In this paper, we propose an improvement of this study by introducing the use of an optimal window of observations. We validate the proposed approach to the Exponentially Weighted Moving Average (EWMA) control charts, which is a commonly used change detection technique. Our strategy is illustrated on a well-known example of the literature. Finally, this windowed entropy-based criterion allows one to design a fault-tolerant fusion methodology, which is experimentally validated from an extended Kalman filter (EKF) in collaborative mobile robotics.

关键词： Microsoft Windows Robots Control charts Change detection algorithms Entropy Fault tolerant systems Collaboration

来源：评论

学校读者我要写书评

暂无评论

Colorization of H&E stained tissue using Deep Learning 40

Colorization of H&E stained tissue using Deep Learning

引用

40th Annual International conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC)

作者： Samsi, Siddharth Jones, Michael Kepner, Jeremy Reuther, Albert MIT Lincoln Lab 244 Wood St Lexington MA 02173 USA

ISBN: (纸本)9781538636466

Histopathology is a critical tool in the diagnosis and stratification of cancer. Digital Pathology involves the scanning of stained and fixed tissue samples to produce high resolution images that can be used for computer-aided diagnosis and research. A common challenge in digital pathology related to the quality and characteristics of staining, which can vary widely from center to center and also within the same institution depending on the age of the stain and other human factors. In this paper we examine the use of deep learning models for colorizing H&E stained tissue images and compare the results with traditional image processing/statistical approaches that have been developed for standardizing or normalizing histopathology images. We adapt existing deep learning models that have been developed for colorizing natural images and compare the results with models developed specifically for digital pathology. Our results show that deep learning approaches can standardize the colorization of H&E images. The performance as measured by the chi-square statistic shows that the deep learning approach can be nearly as good as current state-of-the art normalization methods.

关键词： image color analysis Feature extraction Pathology Training Task analysis Tools

来源：评论

学校读者我要写书评

暂无评论

Towards Evolutionary Super-Resolution 21st

Towards Evolutionary Super-Resolution

引用

21st International conference on the Applications of Evolutionary Computation (EvoApplications)

作者： Kawulok, Michal Benecki, Pawel Kostrzewa, Daniel Skonieczny, Lukasz Future Proc Gliwice Poland Silesian Tech Univ Gliwice Poland

ISBN: (纸本)9783319775388

Super-resolution reconstruction (SRR) allows for producing a high-resolution (HR) image from a set of low-resolution (LR) observations. The majority of existing methods require tuning a number of hyper-parameters which control the reconstruction process and configure the imaging model that is supposed to reflect the relation between high and low resolution. In this paper, we demonstrate that the reconstruction process is very sensitive to the actual relation between LR and HR images, and we argue that this is a substantial obstacle in deploying SRR in practice. We propose to search the hyper-parameter space using a genetic algorithm (GA), thus adapting to the actual relation between LR and HR, which has not been reported in the literature so far. The results of our extensive experimental study clearly indicate that our GA improves the capacities of SRR. Importantly, the GA converges to different values of the hyper-parameters depending on the applied degradation procedure, which is confirmed using statistical tests.

关键词： Genetic algorithm image processing Super-resolution

来源：评论

学校读者我要写书评

暂无评论

Automatic Enhancement of Low-Contrast Monochrome images 38

Automatic Enhancement of Low-Contrast Monochrome Images

引用

IEEE 38th International conference on Electronics and Nanotechnology (ELNANO)

作者： Yelmanov, Sergei Romanyshyn, Yuriy Special Design Off Televis Syst Lvov Ukraine Lviv Polytech Natl Univ Dept Elect & Comp Technol Lvov Ukraine Univ Warmia & Mazury Dept Elect & Comp Technol Olsztyn Poland

ISBN: (纸本)9781538663837

The problem of images enhancement in automatic mode with an acceptable level of computational costs is considered. The task of adaptive enhancement of integral contrast for complex monochrome images on the basis of their nonlinear statistical non-inertial transformations is considered. The research of the effectiveness of the main histogram-based methods of enhancement of the integral contrast of complex images with low-contrast small-sized objects and non-uniform illumination is carried out. A comparative analysis of the effectiveness of no-reference assessing the generalized contrast of complex images using the histogram-based metrics of integral contrast and on the basis of expert assessments is carried out.

关键词： image processing image quality contrast enhancement low-contrast objects automatic mode

来源：评论

学校读者我要写书评

暂无评论

Ultrasound image Despeckling Using stochastic Distance-Based BM3D

引用

IEEE TRANSACTIONS ON image processing 2017年第6期26卷 2632-2643页

作者： Santos, Cid A. N. Martins, Diego L. N. Mascarenhas, Nelson D. A. Univ Fed Sao Carlos Dept Comp BR-13565905 Sao Carlos Brazil Israelite Hosp Albert Einstein Dept Intervent Radiol BR-05651901 Sao Paulo Brazil Fac Campo Limpo Paulista BR-13231230 Campo Limpo Paulista Brazil

Ultrasound image despeckling is an important research field, since it can improve the interpretability of one of the main categories of medical imaging. Many techniques have been tried over the years for ultrasound despeckling, and more recently, a great deal of attention has been focused on patch-based methods, such as non-local means and block-matching collaborative filtering (BM3D). A common idea in these recent methods is the measure of distance between patches, originally proposed as the Euclidean distance, for filtering additive white Gaussian noise. In this paper, we derive new stochastic distances for the Fisher-Tippett distribution, based on well-known statistical divergences, and use them as patch distance measures in a modified version of the BM3D algorithm for despeckling log-compressed ultrasound images. State-of-the-art results in filtering simulated, synthetic, and real ultrasound images confirm the potential of the proposed approach.

关键词： Despeckling ultrasound imaging stochastic distances BM3D patch-based filtering

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：