检索结果-内蒙古大学图书馆

32nd British Machine Vision Conference, BMVC 2021

作者： Wang, Bisheng Fruhwirth-Reisinger, Christian Possegger, Horst Bischof, Horst Cao, Guo School of Computer Science and Engineering Nanjing University of Science and Technology China Christian Doppler Laboratory for Embedded Machine Learning Austria Institute of Computer Graphics and Vision Graz University of Technology Austria

Deep learning methods have led to remarkable progress in multiple object tracking (MOT). However, when tracking in crowded scenes, existing methods still suffer from both inaccurate and missing detections. This paper proposes Detection Refinement for Tracking (DRT) to address these two issues for people tracking. First, we construct an encoder-decoder backbone network with a novel semi-supervised heatmap training procedure, which leverages human heatmaps to obtain a more precise localization of the targets. Second, we integrate a "one patch, multiple predictions" mechanism into DRT which refines the detection results and recovers occluded pedestrians at the same time. Additionally, we leverage a data-driven LSTM-based motion model which can recover lost targets at a negligible computational cost. Compared with strong baseline methods, our DRT achieves significant improvements on publicly available MOT datasets. In addition, DRT generalizes well, i.e. it can be applied to any detector to improve their performance. © 2021. The copyright of this document resides with its authors.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Enhancing textured digital elevation models using photographs 4

Enhancing textured digital elevation models using photograph...

引用

4th International Symposium on 3D Data Processing, Visualization and Transmission, 3DPVT 2008

作者： Schneider, Martin Klein, Reinhard Institute of Computer Science II University of Bonn Römerstrasse 164 Bonn53117 Germany

We present a method for enhancing the visual quality of existing digital elevation models textured with orthophotos, using a sparse set of unordered, high resolution photographs. After an initial manual selection of correspondence points, we automatically register the input photographs to the given terrain data set using robust imagebased modeling techniques. To combine the geo-registered images on the terrain surface, we propose a compositing algorithm that ensures smooth transitions between the images while at the same time preserving the fine details. The resulting textures are inserted into the quadtree representation of a terrain rendering engine to allow an efficient real-time visualization. We demonstrate our method on an HRSC terrain data set and a collection of high resolution photos of Turtmann valley in Switzerland. © 2008 Georgia institute of Technology.

关键词： Geomorphology

来源：评论

学校读者我要写书评

暂无评论

Real-time out-of-core trimmed NURBS rendering and editing

Real-time out-of-core trimmed NURBS rendering and editing

引用

Vision, Modeling and Visualization 2004, VMV 2004

作者： Guthe, Michael Balázs, Ákos Klein, Reinhard Bonn University Institute of Computer Science II Römerstrasse 164 Bonn53117 Germany

ISBN: (纸本)3898380580

For rendering purposes trimmed NURBS surfaces have to be converted into a polygonal representation. In order to fulfill the high quality visualization demands posed by various design and quality control applications, current NURBS rendering methods require a careful preparation of the converted models which often needs manual user intervention. This preprocessing step prevents the user from interactively modifying, removing or adding surfaces during a visualization session. In this paper we present a high quality, out-ofcore trimmed NURBS rendering algorithm that supports both an automatic preprocessing of gigabyte sized models and a real-time rendering of the preprocessed models allowing for the seamless integration of interactive editing of the NURBS surfaces. Additional advantages of our method are the conservative error bounds both for the geometry and the shading, making it suitable even for quality control applications.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Context-based Surface Pattern Completion of Ancient Pottery

Context-based Surface Pattern Completion of Ancient Pottery

引用

20th Eurographics Workshop on graphics and Cultural Heritage, GCH 2022

作者： Lengauer, Stefan Preiner, Reinhold Sipiran, Ivan Karl, Stephan Tinkl, Elisabeth Bustos, Benjamin Schreck, Tobias Graz University of Technology Institute of Computer Graphics and Knowledge Visualization Austria University of Chile Department of Computer Science Chile University of Graz Institute of Classics Austria

ISBN: (纸本)9783038681786

Among various ancient cultures it was common practice to adorn pottery artifacts with lavish surface decoration. While the applied painting styles, color schemes and displayed mythological content may vary greatly, the presence of simple patterns which appear in a repetitive manner can be observed across civilizations and periods. Such pattern sequences generally are arranged in a structured manner in ornament bands or columns that extend over the entire surface of the object. Due to the poor conservation state of many cultural heritage objects, parts of the surface are oftentimes badly damaged or missing altogether. Yet, if the majority of a pattern sequence is preserved, this information can be leveraged to approximate its missing parts. We present an approach that allows the fully automatic determination of the generation rule inherent to a repetitive surface pattern. Based on this generation rule and the preserved patterns from the same pattern class we propose a workflow for reconstruct missing or damaged parts of the surface painting. We evaluate our approach by applying it to a selection of pottery from ancient Peruvian and Greek cultures, showing that our automatic approach is able to handle a variety of problem cases. © ISVR,2023 *** rights reserved

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

GPU-ABiSort: Optimal parallel sorting on stream architectures 20

GPU-ABiSort: Optimal parallel sorting on stream architecture...

引用

20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006

作者： Greß, Alexander Zachmann, Gabriel Institute of Computer Science II Rhein. Friedr.-Wilh.-Universität Bonn Bonn Germany Institute of Computer Science Clausthal University of Technology Clausthal Germany

ISBN: (纸本)1424400546

In this paper, we present a novel approach for parallel sorting on stream processing architectures. It is based on adaptive bitonic sorting. For sorting n values utilizing p stream processor units, this approach achieves the optimal time complexity O((n log n)/p). While this makes our approach competitive with common sequential sorting algorithms not only from a theoretical viewpoint, it is also very fast from a practical viewpoint. This is achieved by using efficient linear stream memory accesses (and by combining the optimal time approach with algorithms optimized for small input sequences). We present an implementation on modern programmable graphics hardware (GPUs). On recent GPUs, our optimal parallel sorting approach has shown to be remarkably faster than sequential sorting on the CPU, and it is also faster than previous non-optimal sorting approaches on the GPU for sufficiently large input sequences. Because of the excellent scalability of our algorithm with the number of stream processor units p (up to n/ log2n or even n/log n units, depending on the stream architecture), our approach profits heavily from, the trend of increasing number of fragment processor units on GPUs, so that we can expect further speed improvement with upcoming GPU generations. © 2006 IEEE.

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Communication pattern based performance prediction on the NCUBE 2 multiprocessor system 3rd

引用

3rd Joint International Conference on Vector and Parallel Processing, CONPAR 1994 - VAPP VI

作者： Wabnig, H. Haring, G. Kranzlmüller, D. Volkert, J. University of Vienna Institute of Applied Computer Science and Information Systems Dept of Advanced Computer Engineering Lenaugasse 2/8 ViennaA-1080 Austria University Linz Institute of Computer Science Dept of Graphics and Parallel Processing Altenbergerstr. 69 LinzA-4040 Austria

ISBN: (纸本)9783540584308

This paper shows the usage of the PAPS toolset for performance prediction on the nCUBE 2 multiprocessor system. Two Petri net models for communication on the nCUBE 2 with different levels of accuracy are developed and investigated. The representation of network contention within the Petri net models is validated for the message broadcast communication pattern. Parameter values for communication are determined for the nCUBE 2 machine. A parallel implementation of the Gaussian elimination algorithm is used to investigate the accuracy of the predicted absolute execution time depending on the choosen Petri net communication model. © Springer-Verlag Berlin Heidelberg 1994.

关键词： Petri nets

来源：评论

学校读者我要写书评

暂无评论

A new integrated neurosymbolic approach for crop-yield prediction using environmental data and satellite imagery at field scale

引用

Artificial Intelligence in Geosciences 2025年第1期6卷

作者： Khadija Meghraoui Teeradaj Racharak Kenza Ait El Kadi Saloua Bensiali Imane Sebari Research Unit of Geospatial Technologies for A Smart Decision. IAV Hassan II Rabat 10101 Morocco Advanced Institute of So-Go-Chi (Convergence Knowledge) Informatics Tohoku University Miyagi Japan School of Information Science Japan Advanced Institute of Science and Technology Ishikawa Japan School of Geomatics and Surveying Engineering. IAV Hassan II Rabat 10101 Morocco Department of Applied Statistics and Computer Science. IAV Hassan II Rabat 10101 Morocco

Crop-yield is a crucial metric in agriculture, essential for effective sector management and improving the overall production process. This indicator is heavily influenced by numerous environmental factors, particularly those related to soil and climate, which present a challenging task due to the complex interactions involved. In this paper, we introduce a novel integrated neurosymbolic framework that combines knowledge-based approaches with sensor data for crop-yield prediction. This framework merges predictions from vectors generated by modeling environmental factors using a newly developed ontology focused on key elements and evaluates this ontology using quantitative methods, specifically representation learning techniques, along with predictions derived from remote sensing imagery. We tested our proposed methodology on a public dataset centered on corn, aiming to predict crop-yield. Our developed smart model achieved promising results in terms of crop-yield prediction, with a root mean squared error (RMSE) of 1.72, outperforming the baseline models. The ontology-based approach achieved an RMSE of 1.73, while the remote sensing-based method yielded an RMSE of 1.77. This confirms the superior performance of our proposed approach over those using single modalities. This integrated neurosymbolic approach demonstrates that the fusion of statistical and symbolic artificial intelligence (AI) represents a significant advancement in agricultural applications. It is particularly effective for crop-yield prediction at the field scale, thus facilitating more informed decision-making in advanced agricultural practices. Additionally, it is acknowledged that results might be further improved by incorporating more detailed ontological knowledge and testing the model with higher-resolution imagery to enhance prediction accuracy.

关键词： Crop-yield prediction Neuro-symbolic AI Ontology Ontology embedding Satellite imagery Machine learning

来源：评论

学校读者我要写书评

暂无评论

iFall - A new embedded system for the detection of unexpected falls

iFall - A new embedded system for the detection of unexpecte...

引用

2010 8th IEEE International Conference on Pervasive Computing and Communications Workshops, PERCOM Workshops 2010

作者： Salomon, Ralf Luder, Marian Bieber, Gerald University of Rostock Faculty of Computer Science and Electrical Engineering 18051 Rostock Germany Fraunhofer-Institute for Computer Graphics Rostock Joachim-Jungius-Strasse 11 18059 Rostock Germany

ISBN: (纸本)9781424466054

This paper describes a new embedded system, called iFall, for the detection of unexpected falls for elderly people. In combination with a new sensors system and the monitoring of its own rotation, iFail is able to detect a fall in its very beginning. It is thus able to initiate some emergency actions, such as inflating a tiny airbag, in order to alleviate a fall's consequences, primarily potential injuries. In that respect, iFail is unique. It's physical properties, such as size, weight, and energy resources, make it almost invisible, so that it can be worn at any time at any occasion by any person. © 2010 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

A VR system for immersive teleoperation and live exploration with a mobile robot

arXiv

引用

arXiv 2019年

作者： Stotko, Patrick Krumpen, Stefan Schwarz, Max Lenz, Christian Behnke, Sven Klein, Reinhard Weinmann, Michael Institute of Computer Science II – Computer Graphics University of Bonn Germany Institute of Computer Science VI – Autonomous Intelligent Systems University of Bonn Germany

— Applications like disaster management and industrial inspection often require experts to enter contaminated places. To circumvent the need for physical presence, it is desirable to generate a fully immersive individual live teleoperation experience. However, standard video-based approaches suffer from a limited degree of immersion and situation awareness due to the restriction to the camera view, which impacts the navigation. In this paper, we present a novel VR-based practical system for immersive robot teleoperation and scene exploration. While being operated through the scene, a robot captures RGB-D data that is streamed to a SLAM-based live multi-client telepresence system. Here, a global 3D model of the already captured scene parts is reconstructed and streamed to the individual remote user clients where the rendering for e.g. head-mounted display devices (HMDs) is performed. We introduce a novel lightweight robot client component which transmits robot-specific data and enables a quick integration into existing robotic systems. This way, in contrast to first-person exploration systems, the operators can explore and navigate in the remote site completely independent of the current position and view of the capturing robot, complementing traditional input devices for teleoperation. We provide a proof-of-concept implementation and demonstrate the capabilities as well as the performance of our system regarding interactive object measurements and bandwidth-efficient data streaming and visualization. Furthermore, we show its benefits over purely video-based teleoperation in a user study revealing a higher degree of situation awareness and a more precise navigation in challenging environments. Copyright © 2019, The Authors. All rights reserved.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

A VR System for Immersive Teleoperation and Live Exploration with a Mobile Robot

A VR System for Immersive Teleoperation and Live Exploration...

引用

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Patrick Stotko Stefan Krumpen Max Schwarz Christian Lenz Sven Behnke Reinhard Klein Michael Weinmann Institute of Computer Science II – Computer Graphics University of Bonn Germany Institute of Computer Science VI – Autonomous Intelligent Systems University of Bonn Germany

Applications like disaster management and industrial inspection often require experts to enter contaminated places. To circumvent the need for physical presence, it is desirable to generate a fully immersive individual live teleoperation experience. However, standard video-based approaches suffer from a limited degree of immersion and situation awareness due to the restriction to the camera view, which impacts the navigation. In this paper, we present a novel VR-based practical system for immersive robot teleoperation and scene exploration. While being operated through the scene, a robot captures RGB-D data that is streamed to a SLAM-based live multiclient telepresence system. Here, a global 3D model of the already captured scene parts is reconstructed and streamed to the individual remote user clients where the rendering for e.g. head-mounted display devices (HMDs) is performed. We introduce a novel lightweight robot client component which transmits robot-specific data and enables a quick integration into existing robotic systems. This way, in contrast to first- person exploration systems, the operators can explore and navigate in the remote site completely independent of the current position and view of the capturing robot, complementing traditional input devices for teleoperation. We provide a proof-of-concept implementation and demonstrate the capabilities as well as the performance of our system regarding interactive object measurements and bandwidth-efficient data streaming and visualization. Furthermore, we show its benefits over purely video-based teleoperation in a user study revealing a higher degree of situation awareness and a more precise navigation in challenging environments.

关键词：

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：