检索结果-内蒙古大学图书馆

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Roberto Souza Richard Frayne Radiology and Clinical Neuroscience University of Calgary Calgary AB Canada Seaman Family MR Research Centre Foothills Medical Centre Calgary AB Canada

Decreasing magnetic resonance (MR) image acquisition times can potentially make MR examinations more accessible. Compressed sensing (CS)-based image reconstruction methods decrease MR acquisition time by reconstructing high-quality images from data that were originally sampled at rates inferior to the Nyquist-Shannon sampling theorem. Deep-learning methods have been used to solve the CS MR reconstruction problem. These proposed methods are able to quickly reconstruct images in a single pass using an appropriately trained network. A variety of different network architectures (e.g., U-nets and Residual U-nets) have been proposed to tackle the CS reconstruction problem. A drawback of these architectures is that they typically only work on image domain data. For undersampled data, the images computed by applying the inverse Fast Fourier Transform (iFFT) are aliased. In this work we propose a hybrid architecture, termed W-net, that works both in the k-space (or frequency-domain) and the image (or spatial) domains. Our network is composed of a complex-valued residual U-net in the k-space domain, an iFFT operation, and a real-valued Unet in the image domain. Our experiments demonstrated, using MR raw k-space data, that the proposed hybrid approach can potentially improve CS reconstruction compared to deep-learning networks that operate only in the image domain. In this study we compare our method with four previously published deep neural networks and examine their ability to reconstruct images that are subsequently used to generate regional volume estimates. Our technique was ranked second in the quantitative analysis, but qualitative analysis indicated that our reconstruction performed the best in hard to reconstruct regions, such as the cerebellum. All images reconstructed with our method were successfully post-processed, and showed good volumetry agreement compared with the fully sampled reconstruction measures.

关键词： image reconstruction Transforms Frequency-domain analysis computer architecture Training Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Deep Painterly Harmonization

引用

computer graphics FORUM 2018年第4期37卷 95-106页

作者： Luan, Fujun Paris, Sylvain Shechtman, Eli Bala, Kavita Cornell Univ Ithaca NY 14853 USA Adobe Res San Jose CA USA

Copying an element from a photo and pasting it into a painting is a challenging task. Applying photo compositing techniques in this context yields subpar results that look like a collage - and existing painterly stylization algorithms, which are global, perform poorly when applied locally. We address these issues with a dedicated algorithm that carefully determines the local statistics to be transferred. We ensure both spatial and inter-scale statistical consistency and demonstrate that both aspects are key to generating quality results. To cope with the diversity of abstraction levels and types of paintings, we introduce a technique to adjust the parameters of the transfer depending on the painting. We show that our algorithm produces significantly better results than photo compositing or global stylization techniques and that it enables creative painterly edits that would be otherwise difficult to achieve.

关键词： CCS Concepts •Computing methodologies → image processing

来源：评论

学校读者我要写书评

暂无评论

Brain Extraction Network Trained with "Silver Standard" Data and Fine-Tuned with Manual Annotation for Improved Segmentation

Brain Extraction Network Trained with "Silver Standard" Data...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Roberto Souza Oeslle Lucena Mariana Bento Julia Garrafa Letícia Rittner Simone Appenzeller Roberto Lotufo Richard Frayne Radiology and Clinical Neurosciences University of Calgary Canada Seaman Family MR Research Centre Foothills Medical Centre Canada Faculty of Life Sciences and Medicine King's College London UK Calgary Image Processing and Analysis Centre Foothills Medical Centre Canada Division of Rheumatology University of Campinas Brazil University of Campinas Campinas Brazil Medical Imaging and Computing Laboratory University of Campinas Brazil

Training convolutional neural networks (CNNs) for medical image segmentation often requires large and representative sets of images and their corresponding annotations. Obtaining annotated images usually requires manual intervention, which is expensive and time consuming, as it typically requires a specialist. An alternative approach is to leverage existing automatic segmentation tools and combine them to create consensus-based "silver-standards" annotations. A drawback to this approach is that silver-standards are usually smooth and this smoothness is transmitted to the output segmentation of the network. Our proposal is to use a two-staged approach. First, silver-standard datasets are used to generate a large set of annotated images in order to train the brain extraction network from scratch. Second, fine-tuning is performed using much smaller amounts of manually annotated data so that the network can learn the finer details that are not preserved in the silver-standard data. As an example, our two-staged brain extraction approach has been shown to outperform seven state-of-the-art techniques across three different public datasets. Our results also suggest that CNNs can potentially capture inter-rater annotation variability between experts who annotate the same set of images following the same guidelines, and also adapt to different annotation guidelines.

关键词： Manuals Standards Silver image segmentation Brain Training Guidelines

来源：评论

学校读者我要写书评

暂无评论

Long-Range Decoder Skip Connections: Exploiting Multi-Context Information for Cardiac image Segmentation

Long-Range Decoder Skip Connections: Exploiting Multi-Contex...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Nicolás Gutierrez-Castilla Ricardo da S. Torres Alexandre X. Falcão Sebastian Kozerke Jürg Schwitter Pier-Giorgio Masci Javier A. Montoya-Zegarra Universidad Catolica San Pablo Arequipa Peru University of Campinas Campinas Brazil Institute for Biomedical Engineering ETH Zurich Zurich Switzerland Lausanne University Hospital Lausanne Switzerland Rayne Institute School of Bioengineering and Imaging Sciences King's College London London United Kingdom Department of Computer Science Universidad Católica San Pablo Arequipa Perú

The heart is one of the most important organs in our body and many critical diseases are associated with its malfunctioning. To assess the risk for heart diseases, Magnetic Resonance Imaging (MRI) has become the golden standard imaging technique, as it provides to the clinicians stacks of images for analyzing the heart structures, such as the ventricles, and thus to make a diagnosis of the patient's health. The problem is that examination of these stacks, often based on the delineation of heart structures, is tedious and error prone due to inter-and intra-variability among manual delineations. For this reason, the investigation of fully automated methods to support heart segmentation is paramount. Most of the successful methods proposed to solve this problem are based on deep-learning solutions. Especially, encoder-decoder architectures, such as the U-Net [1], have demonstrated to be very effective architectures for medical image segmentation. In this paper, we propose to use long-range skip connections on the decoder-part to incorporate multi-context information onto the predicted segmentation masks and also to improve the generalization of the models. In addition, our method obtains smoother segmentations through the combination of feature maps from different stages onto the final prediction layer. We evaluate our approach in the ACDC [2] and LVSC [3] heart segmentation challenges. Experiments performed on both datasets demonstrate that our approach leads to an improvement on both the total Dice score and the Ejection Fraction Correlation, when combined with state-of-the-art encoder-decoder architectures.

关键词： Decoding Heart Three-dimensional displays image segmentation Magnetic resonance imaging Two dimensional displays computer architecture

来源：评论

学校读者我要写书评

暂无评论

Error propagation control in Laplacian mesh compression

引用

computer graphics FORUM 2018年第5期37卷 61-70页

作者： Vasa, L. Dvorak, J. Univ West Bohemia Fac Sci Appl New Technol Informat Soc Plzen Czech Republic Univ West Bohemia Fac Sci Appl Dept Comp Sci & Engn Plzen Czech Republic

Laplacian mesh compression, also known as high-pass mesh coding, is a popular technique for efficiently storing both static and dynamic triangle meshes that gained further recognition with the advent of perceptual mesh distortion evaluation metrics. Currently, the usual rule of thumb that drives the decision for a mesh compression algorithm is whether or not accuracy in absolute scale is required: Laplacian mesh encoding is chosen when perceptual quality is the main objective, while other techniques provide better results in terms of mechanistic error measures such as mean squared error. In this work, we present a modification of the Laplacian mesh encoding algorithm that preserves its benefits while it substantially reduces the resulting absolute error. Our approach is based on analyzing the reconstruction stage and modifying the quantization of differential coordinates, so that the decoded result stays close to the input even in areas that are distant from anchor points. In our approach, we avoid solving an overdetermined system of linear equations and thus reduce data redundancy, improve conditioning and achieve faster processing. Our approach can be directly applied to both static and dynamic mesh compression and we provide quantitative results comparing our approach with the state of the art methods.

关键词： Categories and Subject Descriptors (according to ACM CCS): image processing and computer Vision [I.4.2]: Compression (Coding)—Approximate methods computer graphics [I.3.7]: Animation—

来源：评论

学校读者我要写书评

暂无评论

Applying Hidden Markov Model for Dynamic Game Balancing

Applying Hidden Markov Model for Dynamic Game Balancing

引用

brazilian symposium on Games and Digital Entertainment (SBGAMES)

作者： Marcelo Zamith Jose Ricardo da Silva Junior Esteban W. G. Clua Mark Joselli Instituto Multidisciplinar Universidade Federal Rural do Rio de Janeiro Rio de Janeiro Brazil Instituto Federal do Rio de Janeiro Rio de Janeiro Brazil Instituto de Computação Universidade Federal Fluminense Niteroi Brazil Escola Politécnica Pontifícia Universidade Católica do Paraná Curitiba Brazil

ISBN: (数字)9781728184326

ISBN: (纸本)9781728184333

In Artificial Intelligence (AI) field, Machine Learning (ML) techniques present an interesting approach for games, where it allows some sort of adaptation along the game session. This adaptation can make games more attractive, avoiding that Non-Player-Characters (NPC) present too easy or hard patterns during the game. In both cases, the player may be frustrated due to undesired experience. Although ML techniques are appealing to be used in games, some games characteristics are hard to model. Besides, there are techniques that require a wide variety of observations, which implies two hard barriers for game application: the first is the power processing to compute a huge amount of data in games, considering the real-time characteristic of this kind of application. The second threat is related to the vast majority of games' attributes that must be described in the model. This work proposes a novel approach using ML technique based on Hidden Markov Model (HMM) for game balancing process. HMM is a powerful technique which can be used to learn patterns based on a strong co-relational between an observation and an unknown variable (the hidden part). Our proposed approach learns the player's pattern based on temporal frame observation by co-relating his/her actions (movements) with game events (NPC destruction). The temporal frame observation approach allows the game to learn about player's pattern even if a different person plays it. After the learning process, the following step is to use the knowledge pattern to adapt the game according to the current player, which normally involves making the game harder for a certain period of time. During this time, another pattern may arise, subjected to be learned. In order to validate the presented approach, a Space Invaders clone has been built, allowing to observe that 54 % of participants had more fun while playing it with ML activated in relation to a base version that did not take into account dynamic difficult balancing.

关键词： Games Hidden Markov models Markov processes image color analysis Weapons Tools Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Fast-Forward Methods for Egocentric Videos: A Review

Fast-Forward Methods for Egocentric Videos: A Review

引用

Tutorials of the brazilian symposium on computer graphics and image processing (SIBGRAPI TUTORIALS)

作者： Michel Silva Washington Ramos Alan Neves Edson Araujo Mario Campos Erickson R. Nascimento Department of Computer Science Universidade Federal de Minas Gerais (UFMG) Belo Horizonte Brazil

The emergence of low-cost, high-quality personal wearable cameras combined with a large and increasing storage capacity of video-sharing websites have evoked a growing interest in first-person videos. A First-Person Video is usually composed of monotonous long-running unedited streams captured by a device attached to the user body, which makes it visually unpleasant and tedious to watch. Thus, there is a rise in the need to provide quick access to the information therein. In the last few years, a popular approach to retrieve the information from videos is to produce a short version of the input video by creating a video summary; however, this approach disrupts the temporal context of the recording. Fast-Forward is another approach that creates a shorter version of the video preserving the video context by increasing its playback speed. Although Fast-Forward methods keep the recording story, they do not consider the semantic load of the input video. The Semantic Fast-Forward approach creates a shorter version of First-Person Videos dealing with both video context and emphasis of the relevant portions to keep the semantic load of the input video. In this paper, we present a review of the representative methods in both fast-forward and semantic fast-forward methods and discuss the future directions of the area.

关键词： Videos Cameras Semantics Visualization Hidden Markov models Task analysis Pipelines

来源：评论

学校读者我要写书评

暂无评论

Deep Learning based Exposure Correction for image Exposure Correction with Application in computer Vision for Robotics 15

Deep Learning based Exposure Correction for Image Exposure C...

引用

15th Latin American Robotic symposium (LARS) / 6th brazilian symposium on Robotics (SBR) / 9th Workshop on Robotics in Education (WRE)

作者： Steffens, Cristiano Rafael Jorge Drews-, Paulo Lilles, Jr. Botelho, Silvia Silva Univ Fed Rio Grande FURG Ctr Cilncias Comp C3 Rio Grande Brazil

ISBN: (纸本)9781538677612

This paper presents a convolutional neural network that is able to minimize the effects of poorly exposed digital images in post processing, outperforming state-of-the-art networks on two different conditions: images that are too dark, and images where the irradiance exceeds both the lower and upper bounds of the sensor. The scene's illumination, camera aperture, exposure time, and ISO sensitivity present a direct impact on the quality of the resulting image. Optimize these parameters, however, is a challenging task, which often leads to images that are either too bright, too dark or out of focus. In order to restore over-exposed and under-exposed images, we propose a Generative Adversarial Network (GAN) which learns how to restore an image, preserving and enhancing the contents of the input image, as well as complete regions where data was completely lost due to clipping. Our results show the proposed network is able to enhance color, fulfill regions with missing data, and preserve structural details.

关键词： deep learning image processing image restoration robotics

来源：评论

学校读者我要写书评

暂无评论

Video Segmentation Learning Using Cascade Residual Convolutional Neural Network

Video Segmentation Learning Using Cascade Residual Convoluti...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Daniel F. S. Santos Rafael G. Pires Danilo Colombo João P. Papa Department of Computing São Paulo State University Bauru Brazil Cenpes Petroleo Brasileiro S.A. - Petrobras Rio de Janeiro RJ Brazil São Paulo State University Bauru Brazil

Video segmentation consists of a frame-by-frame selection process of meaningful areas related to foreground moving objects. Some applications include traffic monitoring, human tracking, action recognition, efficient video surveillance, and anomaly detection. In these applications, it is not rare to face challenges such as abrupt changes in weather conditions, illumination issues, shadows, subtle dynamic background motions, and also camouflage effects. In this work, we address such shortcomings by proposing a novel deep learning video segmentation approach that incorporates residual information into the foreground detection learning process. The main goal is to provide a method capable of generating an accurate foreground detection given a grayscale video. Experiments conducted on the Change Detection 2014 and on the private dataset PetrobrasROUTES from Petrobras support the effectiveness of the proposed approach concerning some state-of-the-art video segmentation techniques, with overall F-measures of 0.9535 and 0.9636 in the Change Detection 2014 and PetrobrasROUTES datasets, respectively. Such a result places the proposed technique amongst the top 3 state-of-the-art video segmentation methods, besides comprising approximately seven times less parameters than its top one counterpart.

关键词： Training Gray-scale image segmentation Convolutional neural networks Cameras Meteorology Machine learning

来源：评论

学校读者我要写书评

暂无评论

Modular Latent Spaces for Shape Correspondences

引用

computer graphics FORUM 2018年第5期37卷 199-210页

作者： Ganapathi-Subramanian, Vignesh Diamanti, Olga Guibas, Leonidas J. Stanford Univ Stanford CA 94305 USA Autodesk Inc San Rafael CA USA

We consider the problem of transporting shape descriptors across shapes in a collection in a modular fashion, in order to establish correspondences between them. A common goal when mapping between multiple shapes is consistency, namely that compositions of maps along a cycle of shapes should be approximately an identity map. Existing attempts to enforce consistency typically require recomputing correspondences whenever a new shape is added to the collection, which can quickly become intractable. Instead, we propose an approach that is fully modular, where the bulk of the computation is done on each shape independently. To achieve this, we use intermediate nonlinear embedding spaces, computed individually on every shape;the embedding functions use ideas from diffusion geometry and capture how different descriptors on the same shape inter-relate. We then establish linear mappings between the different embedding spaces, via a shared latent space. The introduction of nonlinear embeddings allows for more nuanced correspondences, while the modularity of the construction allows for parallelizable calculation and efficient addition of new shapes. We compare the performance of our framework to standard functional correspondence techniques and showcase the use of this framework to simple interpolation and extrapolation tasks.

关键词： Categories and Subject Descriptors (according to ACM CCS): 1.3.3 [computer graphics]: Picture/image Generation—Line and curve generation

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：