检索结果-内蒙古大学图书馆

Vertebrae segmentation in 3D CT images based on a variational framework 2

Vertebrae segmentation in 3D CT images based on a variationa...

2nd Workshop on Computational Methods and Clinical Applications for Spine Imaging, CSI 2014 held in conjunction with MICCAI 2014

作者： Hammernik, Kerstin Ebner, Thomas Stern, Darko Urschler, Martin Pock, Thomas Institute for Computer Graphics and Vision BioTechMed Graz University of Technology Graz Austria Ludwig Boltzmann Institute for Clinical Forensic Imaging Graz Austria

ISBN: (纸本)9783319141473

Automatic segmentation of 3D vertebrae is a challenging task in medical imaging. In this paper, we introduce a total variation (TV) based framework that incorporates an a priori model, i.e., a vertebral mean shape, image intensity and edge information. The algorithm was evaluated using leave-one-out cross validation on a data set containing ten computed tomography scans and ground truth segmentations provided for the CSI MICCAI 2014 spine and vertebrae segmentation challenge. We achieve promising results in terms of the Dice Similarity Coefficient (DSC) of 0.93 ± 0.04 averaged over the whole data set. © Springer International Publishing Switzerland 2015.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

A pilot study exploring augmented reality to increase motivation of Chinese college students learning English

引用

computers in Education Journal 2015年第1期6卷 23-33页

作者： Li, Shanshan Chen, Yang Whittinghill, David Vorvoreanu, Mihaela Computer Graphics Technology Department Purdue University United States

With the advent and accelerated development of augmented reality (AR), an increasing number of studies have been conducted to test the effectiveness of this technique in education. Few, however, have investigated how AR might influence students' motivation toward the learning of a second language. To address this gap in the literature, we used a combination of convenience sampling and criterion sampling to select five Chinese college students to evaluate an English vocabulary learning application built upon augmented reality technology. To assess student motivation, the ARCS motivational model was adopted. A semi-structured interview with open-ended questions was used to collect data. Participants indicated that though they were attracted by this tool at the beginning, their motivation level decreased toward the end of the study. An interpretation of our observations in the context of the ARCS model suggests three motivational issues. First, predefined AR materials failed to establish relevance to subjects' personal interests and previous experiences. Secondly, subjects' confidence seemed to have been negatively influenced due to their difficulty in achieving the stated learning objectives. Lastly, technical issues delayed the computer quickly identifying the triggering image and thus resulted in a noticeable lack of system responsiveness. It seems this delay decreased subjects' satisfaction and distracted their attention from the learning task. These factors seemed most determinative in compromising AR's effectiveness as a tool to increase student motivation toward English vocabulary learning. It must be stressed that this study is a pilot with too low number of subjects from which to make any binding generalizations. Nonetheless, these findings should provide useful insights toward the successful application of AR in the educational realm. The authors recommend further study with a larger number of subjects with a wider range of vocabulary samples and a more power

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Improving the affective element in introductory programming coursework for the "non programmer" student

引用

computers in Education Journal 2015年第1期6卷 73-81页

作者： Whittinghill, David M. Nelson, David B. R. Richards, K. Andrew Calahan, Charles A. Computer Graphics Technology Department Purdue University United States

Over a period of several semesters, we examined undergraduate students who were enrolled in an introductory computer-programming course. The goal of the study was to observe the degree to which each student's feelings about the discipline of programming were affected by their experience in this course. The course attempted to encourage a learning environment in which students who were unfamiliar or intimidated by the discipline of programming would be informed that the course is explicitly oriented toward them, rather than toward the more advanced students. The course was designed to defer to the needs of low-skill students such that content progression was slow, thorough, and student centered. Students were surveyed at the beginning of the semester on measures of: self-identified programming skill, years of previous programming experience, and like or dislike of programming. Students were then solicited at the end of semester and surveyed on the degree to which they increased or decreased their enjoyment of programming. As the focus of this approach was oriented toward students with low-positive feelings toward programming, we grouped students into groups of high-positive (HP) and low-positive (LP), and then compared their individual change of attitude toward programming at the end of the semester. We observed that LP students reported greater measures of positive affect toward programming by the end of the semester. These results indicate that approaches to increasing interest in programming education must be accompanied by a supportive, student-centered learning environment that acknowledges the difficulty of the subject matter. © 2015 American Society for Engineering Education. All rights reserved.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Pins 'n' touches: An interface for tagging and editing complex groups 15

Pins 'n' touches: An interface for tagging and editing compl...

引用

10th ACM International Conference on Interactive Tabletops and Surfaces, ITS 2015

作者： Strothoff, Sven Stuerzlinger, Wolfgang Hinrichs, Klaus Visualization and Computer Graphics Research Group University of Münster Germany School of Interactive Arts + Technology Simon Fraser University Vancouver Canada

ISBN: (纸本)9781450338998

Influenced by mouse/pen-based user interfaces, most touchbased object tagging techniques rely mostly on a single interaction point. Once objects are tagged, typically only individual object inclusions/exclusions are possible. Yet, for tagging larger groups with complex spatial layouts, more refinement may be necessary to achieve the desired result. We apply visual tag markers to objects to visualize their group association. Through a new multi-touch pin gesture that "pins" one or more objects "down" while tagging, our new interface is capable of efficiently grouping objects, as identified in our user studies. © Copyright 2015 ACM.

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

Multi-scale 3D rock-art recording

引用

Digital Applications in Archaeology and Cultural Heritage 2015年第2-3期2卷 181-195页

作者： Alexander, Craig Pinz, Axel Reinbacher, Christian McDonald Institute for Archaeological Research University of Cambridge United Kingdom Institute of Electrical Measurement and Measurement Signal Processing Graz University of Technology Austria Institute of Computer Vision and Graphics Graz University of Technology Austria

We present the micro- and mid-scale elements of an integrated multi-scalar solution to the 3D recording of rock-art sites in their landscape contexts. The photogrammetry-based solution integrates 3D models across vastly different scales: individual petroglyph, rock-art panel, site, landscape context. Micro-scale data are acquired with a novel 3D scanner that offers unprecedented radiometric quality at a resolution of 0.1 mm and immediate in-field user feedback while mid-scale data are acquired by unmanned micro-aerial vehicles. At all spatial scales the data are processed in a common Structure-from-Motion environment. Our solution is being tested in Valcamonica in northern Italy. One approximately 400 m² panel - Seradina I Rock 12C - is used throughout the paper to illustrate the approach and results. 3D point clouds are provided as digital supplementary material. It is shown that the solution is well-suited to purpose and will be of value to rock-art archaeology more generally. © 2015 Elsevier Ltd. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bilevel optimization with nonsmooth lower level problems 5

Bilevel optimization with nonsmooth lower level problems

引用

5th International Conference on Scale Space and Variational Methods in computer Vision, SSVM 2015

作者： Ochs, Peter Ranftl, René Brox, Thomas Pock, Thomas Computer Vision Group University of Freiburg Freiburg Germany Institute for Computer Graphics and Vision Graz University of Technology Graz Austria Digital Safety and Security Department AIT Austrian Institute of Technology GmbH Vienna1220 Austria

ISBN: (纸本)9783319184609

We consider a bilevel optimization approach for parameter learning in nonsmooth variational models. Existing approaches solve this problem by applying implicit differentiation to a sufficiently smooth approximation of the nondifferentiable lower level problem. We propose an alternative method based on differentiating the iterations of a nonlinear primal–dual algorithm. Our method computes exact (sub)gradients and can be applied also in the nonsmooth setting. We show preliminary results for the case of multi-label image segmentation. © IFIP International Federation for Information Processing 2015.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Visual Analytics for supporting evidence-based interpretation of molecular cytogenomic findings 15

Visual Analytics for supporting evidence-based interpretatio...

引用

2015 Workshop on Visual Analytics in Healthcare, VAHC 2015

作者： Parsons, Paul Sedig, Kamran Mercer, Robert E. Khordad, Maryam Knoll, Joan Rogan, Peter Computer Graphics Technology Purdue University West LafayetteIN United States Computer Science Western University LondonON Canada Pathology and Laboratory Medicine Western University LondonON Canada Biochemistry Computer Science Western University LondonON Canada

ISBN: (纸本)9781450336710

Interpreting molecular cytogenomic findings that cover the human genome (e.g., microarray results) is challenging, as it requires accessing and working with multiple, diverse sources of data that are often large and heterogeneous. These data need to be accessed, queried, and simultaneously integrated to achieve open-ended goals, such as interpreting findings to make diagnoses and engage in genetic counselling. Currently, typical workows of users are laborious, as data sources are often not integrated and must be accessed separately. Furthermore, large document sets often have to be combed through to assist in interpretation. Analytics tools are needed to help users process and distill large bodies of information into manageable sizes so the most relevant portions can be focused on. Current tools typically do not offer support for interactively exploring and engaging with visual representations of important entities and relationships (e.g., chromosomes, gene-phenotype relationships, and scientific articles). We present VErdICT, a visual analytics tool that can support users in their interpretation of molecular cytogenomic findings. A participatory design approach was taken to make VErdICT human-centered. We describe its development, usability and usefulness, and outline some future research challenges. © 2015 Copyright held by the owner/author(s).

关键词： Genome

来源：评论

学校读者我要写书评

暂无评论

Encoding based Saliency Detection for Videos and Images

Encoding based Saliency Detection for Videos and Images

引用

IEEE Conference on computer Vision and Pattern Recognition

作者： Thomas Mauthner Horst Possegger Georg Waltner Horst Bischof Institute for Computer Graphics and Vision Graz University of Technology

ISBN: (纸本)9781467369657

We present a novel video saliency detection method to support human activity recognition and weakly supervised training of activity detection algorithms. Recent research has emphasized the need for analyzing salient information in videos to minimize dataset bias or to supervise weakly labeled training of activity detectors. In contrast to previous methods we do not rely on training information given by either eye-gaze or annotation data, but propose a fully unsupervised algorithm to find salient regions within videos. In general, we enforce the Gestalt principle of figure-ground segregation for both appearance and motion cues. We introduce an encoding approach that allows for efficient computation of saliency by approximating joint feature distributions. We evaluate our approach on several datasets, including challenging scenarios with cluttered background and camera motion, as well as salient object detection in images. Overall, we demonstrate favorable performance compared to state-of-the-art methods in estimating both ground-truth eye-gaze and activity annotations.

关键词： Videos Image color analysis Encoding Image coding Histograms Training Detectors

来源：评论

学校读者我要写书评

暂无评论

Learning Descriptors for Object Recognition and 3D Pose Estimation

Learning Descriptors for Object Recognition and 3D Pose Esti...

引用

IEEE Conference on computer Vision and Pattern Recognition

作者： Paul Wohlhart Vincent Lepetit Institute for Computer Vision and Graphics Graz University of Technology

ISBN: (纸本)9781467369657

Detecting poorly textured objects and estimating their 3D pose reliably is still a very challenging problem. We introduce a simple but powerful approach to computing descriptors for object views that efficiently capture both the object identity and 3D pose. By contrast with previous manifold-based approaches, we can rely on the Euclidean distance to evaluate the similarity between descriptors, and therefore use scalable Nearest Neighbor search methods to efficiently handle a large number of objects under a large range of poses. To achieve this, we train a Convolutional Neural Network to compute these descriptors by enforcing simple similarity and dissimilarity constraints between the descriptors. We show that our constraints nicely untangle the images from different objects and different views into clusters that are not only well-separated but also structured as the corresponding sets of poses: The Euclidean distance between descriptors is large when the descriptors are from different objects, and directly related to the distance between the poses when the descriptors are from the same object. These important properties allow us to outperform state-of-the-art object views representations on challenging RGB and RGB-D data.

关键词： Training Testing

来源：评论

学校读者我要写书评

暂无评论

In Defense of Color-based Model-free Tracking

In Defense of Color-based Model-free Tracking

引用

IEEE Conference on computer Vision and Pattern Recognition

作者： Horst Possegger Thomas Mauthner Horst Bischof Institute for Computer Graphics and Vision Graz University of Technology

ISBN: (纸本)9781467369657

In this paper, we address the problem of model-free online object tracking based on color representations. According to the findings of recent benchmark evaluations, such trackers often tend to drift towards regions which exhibit a similar appearance compared to the object of interest. To overcome this limitation, we propose an efficient discriminative object model which allows us to identify potentially distracting regions in advance. Furthermore, we exploit this knowledge to adapt the object representation beforehand so that distractors are suppressed and the risk of drifting is significantly reduced. We evaluate our approach on recent online tracking benchmark datasets demonstrating state-of-the-art results. In particular, our approach performs favorably both in terms of accuracy and robustness compared to recent tracking algorithms. Moreover, the proposed approach allows for an efficient implementation to enable online object tracking in real-time.

关键词： Security tracking color representation BENCHMARKS drift drift object information model

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：