检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Pavlo Melnyk Michael Felsberg Mårten Wadenbäck Andreas Robinson Cuong Le Computer Vision Laboratory Department of Electrical Engineering Linköping University Sweden

In this paper, we utilize hyperspheres and regular n-simplexes and propose an approach to learning deep features equivariant under the transformations of nD reflections and rotations, encompassed by the powerful group of O(n). Namely, we propose O(n)-equivariant neurons with spherical decision surfaces that generalize to any dimension n, which we call Deep Equivariant Hyperspheres. We demonstrate how to combine them in a network that directly operates on the basis of the input points and propose an invariant operator based on the relation between two points and a sphere, which as we show, turns out to be a Gram matrix. Using synthetic and real-world data in nD, we experimentally verify our theoretical contributions and find that our approach is superior to the competing methods for O(n)-equivariant benchmark datasets (classification and regression), demonstrating a favorable speed/performance trade-off. The code is available on GitHub.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dependency-aware Task Offloading and Resource Pricing in Vehicular Edge Computing: A Stackelberg Game Approach 22

Dependency-aware Task Offloading and Resource Pricing in Veh...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Zhao, Liang Huang, Shuai Cao, Yuxiang Zhou, Huan Leung, Victor C. M. China Three Gorges University Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering College of Computer and Information Technology Yichang China Northwestern Polytechnical University School of Computer Science Xi'an China The University of British Columbia Department of Electrical and Computer Engineering Vancouver Canada

ISBN: (纸本)9798331509712

Vehicular Edge Computing (VEC) allows vehicles to offload their delay-sensitive tasks to nearby Road Side Units (RSUs) for processing, which improves network quality of service (QoS). However, the self-interested SDN controller is unwilling to ask RSUs to provide free computing resources for vehicles. At the same time, complicated dependencies between vehicular subtasks may cause non-ideal task delay and energy consumption. In order to solve these problems, this paper proposes a Stackelberg game-based Dependency-aware task Offloading and resource Pricing framework (SDOP). Specifically, we first model a vehicular edge network that partially offloads dependency-aware tasks. Then, we depict the interaction between the SDN controller and vehicles as a Stackelberg game, with the goal of maximizing the utility of both parties. Next, we present a Gradient Ascent Plus Genetic algorithm (GAPG) to solve the problem. Finally, numerous simulations are performed, and the results show that compared with other baseline schemes, the proposed GAPG can significantly improve the utility of both the SDN controller and vehicles under various scenarios. © 2024 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Online Self-distillation and Self-modeling for 3D Brain Tumor Segmentation

引用

IEEE Journal of Biomedical and Health Informatics 2025年第6期29卷 4290-4302页

作者： Pang, Yan Li, Yunhao Huang, Teng Liang, Jiaming Wang, Zhen Dong, Changyu Kuang, Dongyang Hu, Ying Chen, Hao Lei, Tim Wang, Qiong The Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China The School of Artificial Intelligence Guangzhou University China The Zhejiang Lab Hangzhou China Sun Yat-sen University China The Department of Computer Science and Engineering The Department of Chemical and Biological Engineering Hong Kong University of Science and Technology China The Department of Electrical Engineering University of Colorado Denver United States

In the specialized domain of brain tumor segmentation, supervised segmentation approaches are hindered by the limited availability of high-quality labeled data, a condition arising from data privacy concerns, significant costs, and ethical issues. In response to this challenge, this paper presents a training framework that adeptly integrates a plug-and-play component, MOD, into current supervised learning models, boosting their efficacy in scenarios with limited data. The MOD consists of an Online Tokenizer and a Dense Predictor, which employs self-distillation and self-modeling on masked patches, promoting swift convergence and efficient representation learning. During the inference phase, the plug-and-play MOD component is excluded, preserving the computational efficiency of the original model without incurring extra processing costs. We substantiated the value of our approach through experiments on leading 3D brain tumor segmentation baselines. Remarkably, models augmented with the MOD consistently showcased superior results, achieving elevated Dice coefficients and HD95 scores on two datasets: BraTS 2021 and MSD 2019 Task-01 Brain Tumor. Code: https://***/aigzhusmart/MOD © 2013 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Pure Detail Feature Extraction Network for Visible-Infrared Re-Identification

引用

Intelligent Automation & Soft Computing 2023年第8期37卷 2263-2277页

作者： Jiaao Cui Sixian Chan Pan Mu Tinglong Tang Xiaolong Zhou Zhejiang University of Technology Hangzhou310023China Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering The College of Computer and InformationChina Three Gorges UniversityYichang443002China The College of Electrical and Information Engineering Quzhou UniversityQuzhou324000China

Cross-modality pedestrian re-identification has important appli-cations in the field of *** to variations in posture,camera per-spective,and camera modality,some salient pedestrian features are difficult to provide effective retrieval ***,it becomes a challenge to design an effective strategy to extract more discriminative pedestrian *** many effective methods for detailed feature extraction are proposed,there are still some shortcomings in filtering background and modality *** further purify the features,a pure detail feature extraction network(PDFENet)is proposed for *** includes three modules,adaptive detail mask generation module(ADMG),inter-detail interaction module(IDI)and cross-modality cross-entropy(CMCE).ADMG and IDI use human joints and their semantic associations to suppress background noise in *** guides the model to ignore modality noise by generating modality-shared feature ***,ADMG generates masks for pedestrian details based on pose *** are used to suppress background information and enhance pedestrian detail ***,IDI mines the semantic relations among details to further refine the ***,CMCE cross-combines classifiers and features to generate modality-shared feature labels to guide model *** ablation experiments as well as visualization results have demonstrated the effectiveness of PDFENet in eliminating background and modality *** addition,comparison experi-ments in two publicly available datasets also show the competitiveness of our approach.

关键词： Person re-identification multimedia image retrieval

来源：评论

学校读者我要写书评

暂无评论

CATPlan: Loss-based Collision Prediction in End-to-End Autonomous Driving

arXiv

引用

arXiv 2025年

作者： Xiong, Ziliang Liu, Shipeng Helgesen, Nathaniel Johnander, Joakim Forssén, Per-Erik Computer Vision Laboratory Department of Electrical Engineering Linköping University Sweden Zenseact Sweden

In recent years, there has been increased interest in the design, training, and evaluation of end-to-end autonomous driving (AD) systems. One often overlooked aspect is the uncertainty of planned trajectories predicted by these systems, despite awareness of their own uncertainty being key to achieve safety and robustness. We propose to estimate this uncertainty by adapting loss prediction from the uncertainty quantification literature. To this end, we introduce a novel light-weight module, dubbed CATPlan, that is trained to decode motion and planning embeddings into estimates of the collision loss used to partially supervise end-to-end AD systems. During inference, these estimates are interpreted as collision risk. We evaluate CATPlan on the safety-critical, nerf-based, closed-loop benchmark NeuroNCAP and find that it manages to detect collisions with a 54.8% relative improvement to average precision over a GMM-based baseline in which the predicted trajectory is compared to the forecasted trajectories of other road users. Our findings indicate that the addition of CATPlan can lead to safer end-to-end AD systems and hope that our work will spark increased interest in uncertainty quantification for such systems. Copyright © 2025, The Authors. All rights reserved.

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

Feature Correlation Aggregation: on the Path to Better Graph Neural Networks

Feature Correlation Aggregation: on the Path to Better Graph...

引用

2023 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2023

作者： Zhou, Jieming Zhang, Tong Fang, Pengfei Petersson, Lars Harandi, Mehrtash Australian National University College of Engineering and Computer Science Canberra Australia Image and Visual Representation Laboratory Epfl Lausanne Switzerland Imaging and Computer Vision Group Csiro Canberra Australia Monash University Department of Electrical and Computer Systems Engineering Melbourne Australia

ISBN: (纸本)9798350382204

Prior to the introduction of Graph Neural Networks (GNNs), modelling and analyzing irregular data, particularly graphs, was thought to be the Achilles' heel of deep learning. The core concept of GNNs is to find a representation by recursively aggregating the representations of a central node and those of its neighbor, and many GNNs' designs have demonstrated its success. However, most only focus on using the first-order information between a node and its neighbors. In this paper, we introduce a central node permutation variant function through a frustratingly simple and innocent-looking modification to the core operation of a GNN, namely the Feature cOrrelation aGgregation (FOG) module, which learns the second-order information from feature correlation between a node and its neighbors in the pipeline. By adding FOG into existing variants of GNNs, we empirically verify this second-order information complements the features generated by original GNNs across a broad set of benchmarks. A tangible boost in the model's performance is observed where the model surpasses previous state-of-the-art results by a significant margin while employing fewer parameters. e.g., 26.202% improvement on a real-world molecular dataset using graph convolutional networks. © 2023 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Quantitative comparison between reflection and transmission modes for the inspection of composite materials using pulsed thermography 45

Quantitative comparison between reflection and transmission ...

引用

Thermosense: Thermal Infrared Applications XLV 2023

作者： Alhammad, M. Avdelidis, N.P. Ibarra-Castanedo, C. Torbali, M.E. Zolotas, A. Maldague, X.P.V. School of Aerospace Transport and Manufacturing Cranfield University United Kingdom Universite Laval Computer Vision and Systems Laboratory Department of Electrical and Computer Engineering Quebec CityG1V 0A6 Canada

ISBN: (纸本)9781510661868

Composite materials have extensively been used in the aerospace industry. There are several inspection methods to ensure the safety of these composites;pulse thermography (PT) is one of the most promising ones. Both reflection and transmission modes of PT could be applied. However, few studies reveal the advantages and disadvantages of these two modes. This paper presents a quantitative comparison of pulsed thermal imaging in two modes of reflection and gradual heat transfer in carbon fibres. Experimental work was conducted on carbon fibre-reinforced plastic (CFRP) samples with different thicknesses and thermal images were recorded in both modes. Thermal images were further processed using statistical analysis and machine learning algorithms. Comparing the results from both modes, there is a marked improvement in the accuracy when the reflection mode is employed. © 2023 SPIE.

关键词： Nondestructive examination

来源：评论

学校读者我要写书评

暂无评论

On Learning Deep O(n)-Equivariant Hyperspheres

arXiv

引用

arXiv 2023年

作者： Melnyk, Pavlo Felsberg, Michael Wadenbäck, Mårten Robinson, Andreas Le, Cuong Computer Vision Laboratory Department of Electrical Engineering Linköping University Sweden

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Exploring Seasonal Variability in the Context of Neural Radiance Fields for 3D Reconstruction on Satellite Imagery 1

引用

46th Annual Conference of the German Association for Pattern Recognition, DAGM-GCPR 2024

作者： Kåreborn, Liv Ingerstad, Erica Berg, Amanda Karlsson, Justus Haglund, Leif Maxar International Sweden AB Ebbegatan 13 Linköping582 13 Sweden AI Sweden Folkungagatan 44 Stockholm118 26 Sweden Computer Vision Laboratory Department of Electrical Engineering Linköping University Linköping581 83 Sweden

ISBN: (数字)9783031851872

ISBN: (纸本)9783031851865

In this work, the seasonal predictive capabilities of Neural Radiance Fields (NeRF) applied to satellite images are investigated. Focusing on the utilization of satellite data, the study explores how Sat-NeRF, a novel approach in computer vision, performs in predicting seasonal variations across different months. Through comprehensive analysis and visualization, the study examines the model’s ability to capture and predict seasonal changes, highlighting specific challenges and strengths. Results showcase the impact of the sun direction on predictions, revealing nuanced details in seasonal transitions, such as snow cover, color accuracy, and texture representation in different landscapes. Given these results, we propose Planet-NeRF, an extension to Sat-NeRF capable of incorporating seasonal variability through a set of month embedding vectors. Comparative evaluations reveal that Planet-NeRF outperforms prior models in the case where seasonal changes are present. The extensive evaluation combined with the proposed method offers promising avenues for future research in this domain. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

GNN-Assisted BiG-AMP: Joint Channel Estimation and Data Detection for Massive MIMO Receiver

引用

IEEE Transactions on Wireless Communications 2025年第6期24卷 4631-4646页

作者： Liu, Zishen Wu, Nan He, Dongxuan Yuan, Weijie Li, Yonghui Quek, Tony Q. S. Beijing Institute of Technology School of Information and Electronics Beijing100081 China Southern University of Science and Technology School of System Design and Intelligent Manufacturing Shenzhen Key Laboratory of Robotics and Computer Vision Shenzhen518055 China The University of Sydney School of Electrical and Information Engineering SydneyNSW2008 Australia Singapore University of Technology and Design Information Systems Technology and Design Pillar Tampines 487372 Singapore Kyung Hee University Department of Electronic Engineering Yongin17104 Korea Republic of

In this paper, we develop a graph neural network (GNN)-assisted bilinear inference approach to enhance the receiver performance of the MIMO system through message passing-based joint channel estimation and data detection (JCD). Specifically, based on the bilinear generalized approximate message passing (BiG-AMP) framework and conditional correlation of signal, we propose a GNN-assisted BiG-AMP (GNN-BiGAMP) approach, which integrates a GNN module into the data-detection-loop to compensate the inaccurate marginal likelihood approximation. By leveraging the coupling between the channel and received symbols, a bilinear GNN-assisted BiG-AMP (BiGNN-BiGAMP) JCD receiver is further proposed. This method incorporates two GNNs with similar graph representation into the bilinear posterior estimation loops, which not only compensates for approximation errors but also alleviates performance loss due to premature variance convergence, thereby enhancing the receiver performance significantly. To fully exploit the supervised information from channel estimation and data detection, we propose a multitask learning based training scheme, which coordinates GNNs with different tasks in two loops. Simulation results show that our proposed GNN-assisted JCD receivers significantly outperform other JCD counterparts in terms of both channel estimation and data detection. © 2002-2012 IEEE.

关键词： Graph neural networks

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：