检索结果-内蒙古大学图书馆

29th British Machine Vision Conference, BMVC 2018

作者： Maurer, Daniel Bruhn, Andres Institute for Visualization and Interactive Systems University of Stuttgart Germany

Temporal coherence is a valuable source of information in the context of optical flow estimation. However, finding a suitable motion model to leverage this information is a non-trivial task. In this paper we propose an unsupervised online learning approach based on a convolutional neural network (CNN) that estimates such a motion model individually for each frame. By relating forward and backward motion these learned models not only allow to infer valuable motion information based on the backward flow, they also help to improve the performance at occlusions, where a reliable prediction is particularly useful. Moreover, our learned models are spatially variant and hence allow to estimate non-rigid motion per construction. This, in turns, allows to overcome the major limitation of recent rigidity-based approaches that seek to improve the estimation by incorporating additional stereo/SfM constraints. Experiments demonstrate the usefulness of our new approach. They not only show a consistent improvement of up to 27% for all major benchmarks (KITTI 2012, KITTI 2015, MPI Sintel) compared to a baseline without prediction, they also show top results for the MPI Sintel benchmark - the one of the three benchmarks that contains the largest amount of non-rigid motion. © 2018. The copyright of this document resides with its authors.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Spatial Partitioning Strategies for Memory-Efficient Ray Tracing of Particles

Spatial Partitioning Strategies for Memory-Efficient Ray Tra...

引用

IEEE Symposium on Large Data Analysis and visualization (LDAV)

作者： Patrick Gralka Ingo Wald Sergej Geringer Guido Reina Thomas Ertl Visualization Research Center University of Stuttgart NVIDIA Corp. Institute for Visualization and Interactive Systems University of Stuttgart

ISBN: (数字)9781728184685

ISBN: (纸本)9781728184692

3D particle data is relevant for a wide range of scientific domains, from molecular dynamics to astrophysics. Simulations in these domains can produce datasets containing millions or billions of particles and rendering needs to be in high quality and interactive to support the scientists in exploring and understanding the structure of their data. One general baseline approach is to represent particles as spheres and employ ray tracing as a rendering technique. However, ray tracing requires the data to be organized in acceleration data structures like bounding volume hierarchies (BVH) to achieve interactive frame rates. Modern GPUs provide hardware acceleration for traversing such data structures but are more limited in memory than CPUs. In this paper, we evaluate different acceleration data structures for sphere-based datasets, including particle kD trees, with respect to their scalability regarding both memory size and speed, and we analyze how these data structures can benefit from hardware acceleration. We show that a bricking of data results in the most effective BVH, both fast to traverse utilizing hardware acceleration and with a reasonably small memory footprint. Additionally, we present a hybrid acceleration data structure that has negligible memory overhead and still ensures reasonable traversal speed. Based on our results, visualization tools and APIs for the ray tracing can provide overall better performance by adapting to the needs of particle-centric application scenarios.

关键词： Acceleration Rendering (computer graphics) Data structures Ray tracing Data visualization Hardware acceleration Pipelines

来源：评论

学校读者我要写书评

暂无评论

Well-being in Isolation: Exploring Artistic Immersive Virtual Environments in a Simulated Lunar Habitat to Alleviate Asthenia Symptoms

arXiv

引用

arXiv 2023年

作者： Pochwatko, Grzegorz Kopec, Wieslaw Swidrak, Justyna Jaskulska, Anna Skorupska, Kinga H. Karpowicz, Barbara Maslyk, Rafal Grzeszczuk, Maciej Barnes, Steven Borkiewicz, Paulina Kobylinski, Pawel Pabiś-Orzeszyna, Michal Balas, Robert Lazarek, Jagoda Dufresne, Florian Bensch, Leonie Nilsson, Tommy Institute of Psychology Polish Academy of Sciences Poland Polish-Japanese Academy of Information Technology Poland Institute of Psychology - PAS Fundació de Recerca Clínic Barcelona Spain IDIBAPS Kobo Association Emotion Cognition Lab SWPS University Poland Visual Narratives Laboratory Lodz Film School Poland National Information Processing Institute Institute of Contemporary Culture University of Lodz Poland Arts et Métiers Institute of Technology Software for Space Systems and Interactive Visualization German Aerospace Center Germany

Revived interest in lunar and planetary exploration is heralding a new era for human spaceflight, characterized by frequent strain on astronaut's mental well-being, which stems from increased exposure to isolated, confined, and extreme (ICE) conditions. Whilst Immersive Virtual Reality (IVR) has been employed to facilitate self-help interventions to mitigate challenges caused by isolated environments in several domains, its applicability in support of future space expeditions remains largely unexplored. To address this limitation, we administered the use of distinct IVR environments to crew members (n=5) partaking in a simulated lunar habitat study. Utilizing a Bayesian approach to scrutinize small group data, we discovered a significant relationship between IVR usage and a reduction in perceived stress-related symptoms, particularly those associated with asthenia (syndrome often linked to chronic fatigue and weakness;a condition characterized by feelings of energy depletion or exhaustion that can be amplified in ICE conditions). The reductions were most prominent with the use of interactive virtual environments. The'Aesthetic Realities' - virtual environments conceived as art exhibits - received exceptional praise from our participants. These environments mark a fascinating convergence of art and science, holding promise to mitigate effects related to isolation in spaceflight training and *** Codes 93B51, 97M50 Copyright © 2023, The Authors. All rights reserved.

关键词： Ecosystems

来源：评论

学校读者我要写书评

暂无评论

Well-being in Isolation: Exploring Artistic Immersive Virtual Environments in a Simulated Lunar Habitat to Alleviate Asthenia Symptoms

Well-being in Isolation: Exploring Artistic Immersive Virtua...

引用

International Symposium on Mixed and Augmented Reality (ISMAR)

作者： Grzegorz Pochwatko Wieslaw Kopec Justyna Swidrak Anna Jaskulska Kinga H. Skorupska Barbara Karpowicz Rafał Masłyk Maciej Grzeszczuk Steven Barnes Paulina Borkiewicz Paweł Kobyliński Michał Pabiś-Orzeszyna Robert Balas Jagoda Lazarek Florian Dufresne Leonie Bensch Tommy Nilsson Polish Academy of Sciences Institute of Psychology Polish-Japanese Academy of Information Technology Fundació de Recerca Clínic Barcelona - IDIBAPS Institute of Psychology - PAS Kobo Association Emotion Cognition Lab SWPS University Visual Narratives Laboratory Lodz Film School National Information Processing Institute Institute of Contemporary Culture University of Lodz Arts et Métiers Institute of Technology Software for Space Systems and Interactive Visualization German Aerospace Center European Space Agency (ESA)

Revived interest in lunar and planetary exploration is heralding a new era for human spaceflight, characterized by frequent strain on astronaut’s mental well-being, which stems from increased exposure to isolated, confined, and extreme (ICE) conditions. Whilst Immersive Virtual Reality (IVR) has been employed to facilitate self-help interventions to mitigate challenges caused by isolated environments in several domains, its applicability in support of future space expeditions remains largely unexplored. To address this limitation, we administered the use of distinct IVR environments to crew members $(n=5)$ partaking in a simulated lunar habitat study. Utilizing a Bayesian approach to scrutinize small group data, we discovered a significant relationship between IVR usage and a reduction in perceived stress-related symptoms, particularly those associated with asthenia (syndrome often linked to chronic fatigue and weakness; a condition characterized by feelings of energy depletion or exhaustion that can be amplified in ICE conditions). The reductions were most prominent with the use of interactive virtual environments. The ’Aesthetic Realities’ - virtual environments conceived as art exhibits - received exceptional praise from our participants. These environments mark a fascinating convergence of art and science, holding promise to mitigate effects related to isolation in spaceflight training and beyond

关键词：

来源：评论

学校读者我要写书评

暂无评论

Subjective annotation for a frame interpolation benchmark using artefact amplification

引用

Quality and User Experience 2020年第1期5卷 1-18页

作者： Men, Hui Hosu, Vlad Lin, Hanhe Bruhn, Andrés Saupe, Dietmar Department of Computer and Information Science University of Konstanz Konstanz Germany Institute for Visualization and Interactive Systems University of Stuttgart Stuttgart Germany

Current benchmarks for optical flow algorithms evaluate the estimation either directly by comparing the predicted flow fields with the ground truth or indirectly by using the predicted flow fields for frame interpolation and then comparing the interpolated frames with the actual frames. In the latter case, objective quality measures such as the mean squared error are typically employed. However, it is well known that for image quality assessment, the actual quality experienced by the user cannot be fully deduced from such simple measures. Hence, we conducted a subjective quality assessment crowdscouring study for the interpolated frames provided by one of the optical flow benchmarks, the Middlebury benchmark. It contains interpolated frames from 155 methods applied to each of 8 contents. For this purpose, we collected forced-choice paired comparisons between interpolated images and corresponding ground truth. To increase the sensitivity of observers when judging minute difference in paired comparisons we introduced a new method to the field of full-reference quality assessment, called artefact amplification. From the crowdsourcing data (3720 comparisons of 20 votes each) we reconstructed absolute quality scale values according to Thurstone’s model. As a result, we obtained a re-ranking of the 155 participating algorithms w.r.t. the visual quality of the interpolated frames. This re-ranking not only shows the necessity of visual quality assessment as another evaluation metric for optical flow and frame interpolation benchmarks, the results also provide the ground truth for designing novel image quality assessment (IQA) methods dedicated to perceptual quality of interpolated images. As a first step, we proposed such a new full-reference method, called WAE-IQA, which weights the local differences between an interpolated image and its ground truth.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improving natural language processing tasks with human gaze-guided neural attention 20

Improving natural language processing tasks with human gaze-...

引用

Proceedings of the 34th International Conference on Neural Information Processing systems

作者： Ekta Sood Simon Tannert Philipp Müller Andreas Bulling University of Stuttgart Institute for Visualization and Interactive Systems (VIS) Germany University of Stuttgart Institute for Natural Language Processing (IMS) Germany

ISBN: (纸本)9781713829546

A lack of corpora has so far limited advances in integrating human gaze data as a supervisory signal in neural attention mechanisms for natural language processing (NLP). We propose a novel hybrid text saliency model (TSM) that, for the first time, combines a cognitive model of reading with explicit human gaze supervision in a single machine learning framework. On four different corpora we demonstrate that our hybrid TSM duration predictions are highly correlated with human gaze ground truth. We further propose a novel joint modeling approach to integrate TSM predictions into the attention layer of a network designed for a specific upstream NLP task without the need for any task-specific human gaze data. We demonstrate that our joint model outperforms the state of the art in paraphrase generation on the Quora Question Pairs corpus by more than 10% in BLEU-4 and achieves state of the art performance for sentence compression on the challenging Google Sentence Compression corpus. As such, our work introduces a practical approach for bridging between data-driven and cognitive models and demonstrates a new way to integrate human gaze-guided neural attention into NLP tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Subjective annotation for a frame interpolation benchmark using artifact amplification

arXiv

引用

arXiv 2020年

作者： Men, Hui Hosu, Vlad Lin, Hanhe Bruhn, Andrés Saupe, Dietmar Department of Computer and Information Science University of Konstanz Germany Institute for Visualization and Interactive Systems University of Stuttgart Germany

Current benchmarks for optical flow algorithms evaluate the estimation either directly by comparing the predicted flow fields with the ground truth or indirectly by using the predicted flow fields for frame interpolation and then comparing the interpolated frames with the actual frames. In the latter case, objective quality measures such as the mean squared error are typically employed. However, it is well known that for image quality assessment, the actual quality experienced by the user cannot be fully deduced from such simple measures. Hence, we conducted a subjective quality assessment crowdscouring study for the interpolated frames provided by one of the optical flow benchmarks, the Middlebury benchmark. It contains interpolated frames from 155 methods applied to each of 8 contents. For this purpose, we collected forced choice paired comparisons between interpolated images and corresponding ground truth. To increase the sensitivity of observers when judging minute difference in paired comparisons we introduced a new method to the field of full-reference quality assessment, called artifact amplification. From the crowdsourcing data (5887 participants, 3720 comparisons of 50 votes each) we reconstructed absolute quality scale values according to Thurstone's model. As a result, we obtained a re-ranking of the 155 participating algorithms w.r.t. the visual quality of the interpolated frames. This re-ranking not only shows the necessity of visual quality assessment as another evaluation metric for optical flow and frame interpolation benchmarks, the results also provide the ground truth for designing novel image quality assessment (IQA) methods dedicated to perceptual quality of interpolated images. As a first step, we proposed such a new full-reference method, called WAE-IQA. By weighing the local differences between an interpolated image and its ground truth WAE-IQA performed slightly better than the currently best FR-IQA approach from the literature. Copyright ©

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Integrative computational design and construction: Rethinking architecture digitally

Civil Engineering Design

引用

Civil Engineering Design 2021年第4期3卷 123-135页

作者： Knippers, Jan Kropp, Cordula Menges, Achim Sawodny, Oliver Weiskopf, Daniel Institute for Building Construction and Structural Design University of Stuttgart Stuttgart Germany Institute for Social Sciences–Chair of Sociology of Technology Risk and Environment University of Stuttgart Stuttgart Germany Institute for Computational Design and Construction University of Stuttgart Stuttgart Germany Institute for System Dynamics University of Stuttgart Stuttgart Germany Institute for Visualization and Interactive Systems University of Stuttgart Stuttgart Germany

Increasing the construction capacity, while at the same time significantly reducing harmful emissions and consumption of nonrenewable resources, and still providing a liveable and affordable built environment, provides a great challenge for future construction. In order to achieve this, both the productivity of construction processes and the energy and resource efficiency of construction systems have to be improved in a reciprocal process. Digital technologies make it possible to address these challenges in novel ways. The vision of this Cluster of Excellence IntCDC at the University of Stuttgart and the Max Planck institute for Intelligent systems is to harness the full potential of digital technologies to rethink design and construction based on integration and interdisciplinarity, with the goal of laying the methodological foundations to profoundly modernize the design and construction process and related building systems by adopting a systematic, holistic and integrative computational approach. One key objective is to develop an overarching methodology of "co-designing" methods, processes and systems based on interdisciplinary research encompassing architecture, structural engineering, building physics, engineering geodesy, manufacturing and systems engineering, computer science and robotics, and humanities and social sciences. In this way, the Cluster aims to address the ecological, economic and social challenges and to provide the prerequisites for a high-quality and sustainable built environment and a digital building culture. © 2021 The Authors. Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG, Berlin

关键词： Intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Pattern-Based Semantic and Temporal Exploration of Social Media Messages

Pattern-Based Semantic and Temporal Exploration of Social Me...

引用

IEEE Symposium on Visual Analytics Science and Technology (VAST)

作者： Johannes Knittel Steffen Koch Thomas Ertl Institute of Visualization and Interactive Systems University of Stuttgart

Social media is a valuable source for emergency workers and first-responders due to its wide use. Unfortunately, the number of posts poses a challenge to quickly find relevant and trustworthy information. We propose a text pattern-based approach to gain insights from large micro-document collections, including temporal and semantic relationships. In this work, we apply our method to the fictional data set of the VAST 2019 Mini-Challenge 3 that deals with the aftermath of an earthquake. We present major findings we could deduct using our visual analytics approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LabelTransfer-Integrating Static and Dynamic Label Representation for Focus+Context Text Exploration

LabelTransfer-Integrating Static and Dynamic Label Represent...

引用

2018 International Symposium on Big Data Visual and Immersive Analytics, BDVA 2018

作者： Han, Qi John, Markus Koch, Steffen Assenov, Ivan Ertl, Thomas Institute for Visualization and Interactive Systems University of Stuttgart Stuttgart Germany

ISBN: (纸本)9781538691946

In recent years, interactive visualization to analyze text documents has gained an impressive momentum. This is not surprising considering the fast increase of electronically available textual documents of various kinds. These include, for example, patents, scholarly documents, social media messages, and many other sources that contain valuable knowledge and insights for many stakeholders. interactive text visualization turned out to be an important means for exploring and gaining insights into complex and often large document collections. An established visualization strategy to represent such collections is using projection-based techniques that visualize documents as glyphs in a 2D view aiming to reflect the semantic similarity of documents by the proximity of their placement. Static labels have been suggested to characterize the overall topics contained in the projected data to improve the effectiveness of such visualization techniques. Other approaches employ magic lenses that enable users to explore the 2D spatialization freely on various granularity levels. In this work, we propose a visual exploration approach that combines cluster-based labeling of projected documents with an interaction concept for magic lens techniques. We offer a set of novel interactive features to support a smooth transition between static labels and the magic lens approach while exploiting the different levels of visual abstraction of both techniques without introducing additional clutter through overdraw. Finally, we provide insights gained from a preliminary user study and present the benefits of our approach. © 2018 IEEE.

关键词： visualization

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：