检索结果-内蒙古大学图书馆

2018 colour and visual computing Symposium, CVCS 2018

作者： Valeriano, Leonel Cuevas Thomas, Jean-Baptiste Benoit, Alexandre Norwegian Colour and Visual Computing Laboratory NTNU Gjovik Norway LISTIC Univ. Savoie Mont Blanc Annecy France

ISBN: (纸本)9781538656457

We compare a recent dehazing method based on deep learning, Dehazenet, with traditional state-of-the-art approaches, on benchmark data with reference. Dehazenet estimates the depth map from transmission factor on a single color image, which is used to inverse the Koschmieder model of imaging in the presence of haze. In this sense, the solution is still attached to the Koschmieder model. We demonstrate that the transmission is very well estimated by the network, but also that this method exhibits the same limitation than others due to the use of the same imaging model. © 2018 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Multiscale daltonization in the gradient domain 26

Multiscale daltonization in the gradient domain

引用

26th Color and Imaging Conference: Color Science and Engineering Systems, Technologies, and Applications, CIC 2018

作者： Simon-Liedtke, Joschua Thomas Farup, Ivar Norwegian Colour and Visual Computing Laboratory Department of Computer Science NTNU-Norwegian University of Science and Technology Gjøvik Norway

ISBN: (纸本)9780892083374

We propose a daltonization method that enhances chromatic edges and contrast for color-deficient people by optimizing the gradient of an image. We rotate and scale the error gradient between the original and its simulation in the color space into the direction of optimal visibility that is orthogonal to both the main direction of information loss and the direction of lightness. Then, we reintegrate the daltonized image version from the modified gradient through an iterative diffusion process. Moreover, we include multiscaling to guarantee optimal daltonization on different scales of the image. Also, we provide an interface for data attachment modules designed to maintain naturalness of memory colors like neutral colors. We evaluate and compare our proposed method to other top-performing daltonization methods in behavioral and psychometric experiments. A visual-search experiment assessing performance of the attentional mechanism of the human visual system before and after daltonization measures the greatest improvement in accuracy for our proposed method compared to the original and all investigated daltonization methods. It also reveals optimal results for both natural and Ishihara images among both protan and deutan color-deficient observers. Furthermore, we can deduce from the results of a pairwise preference evaluation that our proposed method is preferred highest amongst all color-deficient people in total. Our proposed method is also ranked among the most preferred daltonization methods for both protan and deutan color-deficient observers individually. © 2018, Society for Imaging Science and Technology.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Assessing colour differences under a wide range of luminance levels using surface and display colours 27

Assessing colour differences under a wide range of luminance...

引用

27th Color and Imaging Conference: Color Science and Engineering Systems, Technologies, and Applications, CIC 2019

作者： Xu, Qiang Cui, Guihua Safdar, Muhammad Xu, Lihao Luo, Ming Ronnier Zhao, Baiyue State Key Laboratory of Modern Optical Instrumentation Zhejiang University Hangzhou China School of Physics & Electronics Information Engineering Wenzhou University Wenzhou China Norwegian Colour and Visual Computing Laboratory NTNU Norwegian University of Science and Technology Gjovik Norway School of Digital Media and Art Design Hangzhou Dianzi University Hangzhou China School of Design Leeds University Leeds United Kingdom

ISBN: (纸本)9780892083435

Two experimental data sets were accumulated for evaluating colour differences under a wide range of luminance levels using the printed colours in a spectrum tuneable viewing cabinet and selfluminous colours on a display respectively. For the surface mode experiment, pairs of samples were assessed at 9 phases ranged from 0.25 to 1128 cd/m2. For the luminous mode experiment, it was conducted in 6 phases ranged from 0.25 to 279 cd/m2. There were 140 and 42 pairs of samples judged by 20 observers using a sixcategory scales for each phase. The results were used to establish the just noticeable difference (JND) at each luminance level and showed a great agreement between two modes of colours. These were used to test the performance of 5 uniform colour spaces and colour difference equations. Also, the spaces or equations were extended to improve the fit to the present data sets. © 2019 Society for Imaging Science and Technology.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Quality Assessment of Reconstruction and Relighting from RTI Images: Application to Manufactured Surfaces

Quality Assessment of Reconstruction and Relighting from RTI...

引用

International IEEE Conference on Signal-Image Technologies and Internet-Based System

作者： Abir Zendagui Jean-Baptiste Thomas Gaëtan Le Goic Yuly Castro Marvin Nurit Alamin Mansouri Marius Pedersen Imagerie et Vision Artificielle Laboratory UFR Sciences et Techniques University of Burgundy Dijon France The Norwegian Colour and Visual Computing Laboratory NTNU Gjøvik Norway University of Burgundy Dijon France

ISBN: (数字)9781728156866

ISBN: (纸本)9781728156873

In this paper, we propose to evaluate the quality of the reconstruction and relighting from images acquired by a Reflectance Transformation Imaging (RTI) device. Three relighting models, namely the PTM, HSH and DMD, are evaluated using PSNR and SSIM. A visual assessment of how the reconstructed surfaces are perceived is also carried out through a sensory experiment. This study allows to estimate the relevance of these models to reproduce the appearance of the manufactured surfaces. It also shows that DMD reproduces the most accurate reconstruction/relighting to an acquired measurement and that a higher sampling density don't mean necessarily a higher perceptual quality.

关键词： Image reconstruction Surface reconstruction Surface treatment Surface topography Harmonic analysis Imaging

来源：评论

学校读者我要写书评

暂无评论

DEEP SMOKE REMOVAL FROM MINIMALLY INVASIVE SURGERY VIDEOS

DEEP SMOKE REMOVAL FROM MINIMALLY INVASIVE SURGERY VIDEOS

引用

IEEE International Conference on Image Processing

作者： Sabri Bolkar Congcong Wang Faouzi Alaya Cheikh Sule Yildirim Norwegian Colour and Visual Computing Laboratory NTNU Norway

During video-guided minimally invasive surgery, quality of frames may be degraded severely by cauterization-induced smoke and condensation of vapor. This degradation of quality creates discomfort for the operating surgeon, and causes serious problems for automatic follow-up processes such as registration, segmentation and tracking. This paper proposes a novel deep neural network based smoke removal solution that is able to enhance the quality of surgery video frames in real-time. It employs synthetically generated training dataset including smoke embedded and clean reference versions. Results calculated on the test set indicate that our network outperforms previous defogging methods in terms of quantitative and qualitative measures. While eliminating apparent smoke, it also successfully preserves the natural appearance of tissue surface. To the best of our knowledge, the presented method is the first deep neural network based approach for the surgical field smoke removal problem.

关键词： Videos Training Mathematical model Image color analysis Neural networks Minimally invasive surgery

来源：评论

学校读者我要写书评

暂无评论

Novel Approach to Uniformization of a Color Space via Generic Deep Learning-Based Transformation

Novel Approach to Uniformization of a Color Space via Generi...

引用

colour and visual computing Symposium

作者： Oleksii Sidorov The Norwegian Colour and Visual Computing Laboratory NTNU Gjovik Norway

Creation of a convenient uniform color space which will match experimental data and will be easy-to-use - one of a primary objectives of color science. In this work we present novel approach which transform any existing color space in order to make the discrimination ellipses unit circles. This allows to find color difference in a new color space as Euclidian distance with constant unit threshold. A priori, color difference in a new color space should match perceptual color difference, since transformation is based on experimental data. The lack of experimental data for every point may be overcame via deep learning-based regression model, namely, multilayer perceptron which takes color coordinates as input and returns ellipses' parameters as output. Results are demonstrated on MacAdam's ellipses in xy Chromaticity Diagram and Luo and Rigg chromaticity-discrimination ellipses in a*b* chromaticity plane. Additionally, topic of fitting discrimination ellipses is covered.

关键词： Image color analysis Fitting Multilayer perceptrons Bayes methods Optimization Training visualization

来源：评论

学校读者我要写书评

暂无评论

Web Technologies Enable Agile Color Management

Web Technologies Enable Agile Color Management

引用

International IEEE Conference on Signal-Image Technologies and Internet-Based System

作者： Philippe Colantoni Jean-Baptiste Thomas Alain Trémeau Jon Yngve Hardeberg Laboratoire Hubert Curien UMR Université de Lyon Université Jean Monnet The Norwegian Colour and Visual Computing Laboratory NTNUNorwegian University of Science and Technology Université de Lyon Université Jean Monnet

ISBN: (数字)9781728156866

ISBN: (纸本)9781728156873

With the number of display technologies, cameras, operating systems and software solutions, one of the only technologies that is compatible across this diversity is the web browser. We propose to show that the technologies now present in web browsers allow an indepent management of the color information on large variety of devices. For this purpose we introduce the basic concepts of color management and then we show how to implement them with WebAssembly and WebGL by introducing the concept of WebCMM. A WebCMM adapted for the color management of HTML elements in 2D, 3D but also for virtual environments. Finally, we present how we can implement this WebCMM for a real case of color workflow implemented in a demonstrator web page.

关键词： Image color analysis Browsers Three-dimensional displays Rendering (computer graphics) Software Standards

来源：评论

学校读者我要写书评

暂无评论

Artificial Color Constancy via GoogLeNet with Angular Loss Function

arXiv

引用

arXiv 2018年

作者： Sidorov, Oleksii Norwegian Colour and Visual Computing Laboratory NTNU Gjovik Norway

Color Constancy is the ability of the human visual system to perceive colors unchanged independently of the illumination. Giving a machine this feature will be beneficial in many fields where chromatic information is used. Particularly, it significantly improves scene understanding and object *** this paper, we propose transfer learning-based algorithm, which has two main features: accuracy higher than many state-of-the-art algorithms and simplicity of implementation. Despite the fact that GoogLeNet was used in the experiments, given approach may be applied to any CNN. Additionally, we discuss design of a new loss function oriented specifically to this problem, and propose a few the most suitable options. Copyright © 2018, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Conditional GANs for Multi-Illuminant Color Constancy: Revolution or Yet Another Approach?

arXiv

引用

arXiv 2018年

作者： Sidorov, Oleksii Norwegian Colour and Visual Computing Laboratory NTNU Gjøvik Norway

Non-uniform and multi-illuminant color constancy are important tasks, the solution of which will allow to discard information about lighting conditions in the image. Nonuniform illumination and shadows distort colors of real-world objects and mostly do not contain valuable information. Thus, many computer vision and image processing techniques would benefit from automatic discarding of this information at the pre-processing step. In this work we propose novel view on this classical problem via generative end-to-end algorithm based on image conditioned Generative Adversarial Network. We also demonstrate the potential of the given approach for joint shadow detection and removal. Forced by the lack of training data, we render the largest existing shadow removal dataset and make it publicly available. It consists of approximately 6,000 pairs of wide field of view synthetic images with and without shadows. Copyright © 2018, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Changing the Image Memorability: From Basic Photo Editing to GANs

arXiv

引用

arXiv 2018年

作者： Sidorov, Oleksii Norwegian Colour and Visual Computing Laboratory NTNU Gjøvik Norway

Memorability is considered to be an important characteristic of visual content, whereas for advertisement and educational purposes it is often crucial. Despite numerous studies on understanding and predicting image memorability, there are almost no achievements in memorability modification. In this work, we study two approaches to image editing – GAN and classical image processing – and show their impact on memorability. The visual features which influence memorability directly stay unknown till now, hence it is impossible to control it manually. As a solution, we let GAN learn it deeply using labeled data, and then use it for conditional generation of new images. By analogy with algorithms which edit facial attributes, we consider memorability as yet another attribute and operate with it in the same way. Obtained data is also interesting for analysis, simply because there are no real-world examples of successful change of image memorability while preserving its other attributes. We believe this may give many new answers to the question "what makes an image memorable?" Apart from that we also study the influence of conventional photo-editing tools (Photoshop, Instagram, etc.) used daily by a wide audience on memorability. In this case, we start from real practical methods and study it using statistics and recent advances in memorability prediction. Photographers, designers, and advertisers will benefit from the results of this study directly. Copyright © 2018, The Authors. All rights reserved.

关键词： Image processing

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：