检索结果-内蒙古大学图书馆

TECMH:Transformer-Based Cross-Modal Hashing For Fine-Grained Image-Text Retrieval

computers, Materials & Continua 2023年第5期75卷 3713-3728页

作者： Qiqi Li Longfei Ma Zheng Jiang Mingyong Li Bo Jin College of Computer and Information Science Chongqing Normal UniversityChongqing401331China Institute of Systems and Robotics(ISR) Department of Electrical and Computer Engineering(DEEC)University of CoimbraCoimbraPortugal

In recent years,cross-modal hash retrieval has become a popular research field because of its advantages of high efficiency and low ***-modal retrieval technology can be applied to search engines,crossmodalmedical processing,*** existing main method is to use amulti-label matching paradigm to finish the retrieval ***,such methods do not use fine-grained information in the multi-modal data,which may lead to suboptimal *** avoid cross-modal matching turning into label matching,this paper proposes an end-to-end fine-grained cross-modal hash retrieval method,which can focus more on the fine-grained semantic information of multi-modal ***,the method refines the image features and no longer uses multiple labels to represent text features but uses BERT for ***,this method uses the inference capabilities of the transformer encoder to generate global fine-grained ***,in order to better judge the effect of the fine-grained model,this paper uses the datasets in the image text matching field instead of the traditional label-matching *** article experiment on Microsoft COCO(MS-COCO)and Flickr30K datasets and compare it with the previous *** experimental results show that this method can obtain more advanced results in the cross-modal hash retrieval field.

关键词： Deep learning cross-modal retrieval hash learning transformer

来源：评论

学校读者我要写书评

暂无评论

Multispectral Image Segmentation in Agriculture: Evaluating Deep Learning Models with Train-Test Split and Cross-Validation Strategies

Multispectral Image Segmentation in Agriculture: Evaluating ...

引用

Iberian robotics Conference (ROBOT)

作者： Wilgo Cardoso Tiago Barros Gil Gonçalves Cristiano Premebida Urbano J. Nunes Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Portugal Dept. of Mathematics Institute for Systems Engineering and Computers at Coimbra University of Coimbra Portugal

ISBN: (数字)9798350376364

ISBN: (纸本)9798350376371

In agricultural robotics, the integration of multi-spectral image processing and deep learning (DL) has become the state-of-the-art (SOTA) in crop monitoring, yield estimation, and efficient land management. This work addresses the impact of different DL-segmentation models and evaluation protocols on multispectral imagery datasets collected by a UAV over vineyards. In terms of evaluation protocols, we have considered train-test split, standard k-fold cross-validation, and group k-fold cross-validation. While the first two assume that the training and test data are drawn from the same underlying distribution, the group k-fold cross-validation protocol assumes that each fold represents distinct distributions. Most works either adopt train-test split or k-fold cross-validation under the assumption that both the training and test sets are drawn from the same distribution. However, this assumption is rarely met in real-world applications. Therefore, the objective of this study is to evaluate and compare different evaluation protocols within the context of a real-world agricultural task, highlighting their limitations and weaknesses. Two SOTA DL-based segmentation models, SegNet and DeepLabV3, are employed to perform semantic segmentation on datasets of three Vineyards. The models have been trained and tested considering single-modality representations. In addition to the RGB modality, models trained on NDVI, GNDVI and early fusion are also evaluated. The performance of the models are evaluated using the IoU metric across different dataset configurations. The results indicate that the early fusion representation achieves the highest performance across the various splitting protocols, compared to the single-input representations. The results also show that the train-test and random k-fold splitting approaches report similar results. However, when employing group k-fold the performance drops consistently across both models and the modalities. This indicates that the models

关键词： Deep learning Training Precision agriculture Measurement Protocols Data models Yield estimation Robots Monitoring Standards

来源：评论

学校读者我要写书评

暂无评论

Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification

Strength in Diversity: Multi-Branch Representation Learning ...

引用

International Conference on Intelligent Transportation

作者： Eurico Almeida Bruno Silva Jorge Batista Institute of Systems and Robotics University of Coimbra Coimbra Portugal Department of Electrical and Computers Engineering Faculty of Science and Technology University of Coimbra Coimbra Portugal

This paper presents an efficient and lightweight multi-branch deep architecture to improve vehicle reidentification (V-ReID). While most V-ReID work uses a combination of complex multi-branch architectures to extract robust and diversified embeddings towards re-identification, we advocate that simple and lightweight architectures can be designed to fulfill the Re-Idtask without compromising performance. We propose a combination of Grouped-convolution and Loss-Branch-Split strategies to design a multi-branch architecture that improve feature diversity and feature discriminability. We combine a ResNet50 global branch architecture with a BotNet self-attention branch architecture, both designed within a Loss-Branch-Split (LBS) strategy. We argue that specialized loss-branch-splitting helps to improve re-identification tasks by generating specialized re-identification features. A lightweight solution using grouped convolution is also proposed to mimic the learning of loss-splitting into multiple embeddings while significantly reducing the model size. In addition, we designed an improved solution to leverage additional metadata, such as camera ID and pose information, that uses 97% less parameters, further improving re-identification performance. In comparison to state-of-the-art (SoTA) methods, our approach outperforms competing solutions in Veri-776 by achieving 85.6% mAP and 97.7% CMC1 and obtains competitive results in Veri-Wild with 88.1% mAP and 96.3% CMC1. Overall, our work provides important insights into improving vehicle re-identification and presents a strong basis for other retrieval tasks. Our code is available at the link.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification

arXiv

引用

arXiv 2023年

作者： Almeida, Eurico Silva, Bruno Batista, Jorge The Institute of Systems and Robotics University of Coimbra Coimbra3030-219 Portugal The Department of Electrical and Computers Engineering Faculty of Science and Technology University of Coimbra Coimbra3030-290 Portugal

This paper presents an efficient and lightweight multi-branch deep architecture to improve vehicle re-identification (V-ReID). While most V-ReID work uses a combination of complex multi-branch architectures to extract robust and diversified embeddings towards re-identification, we advocate that simple and lightweight architectures can be designed to fulfill the Re-ID task without compromising performance. We propose a combination of Grouped-convolution and Loss-Branch-Split strategies to design a multi-branch architecture that improve feature diversity and feature discriminability. We combine a ResNet50 global branch architecture with a BotNet self-attention branch architecture, both designed within a Loss-Branch-Split (LBS) strategy. We argue that specialized loss-branch-splitting helps to improve re-identification tasks by generating specialized re-identification features. A lightweight solution using grouped convolution is also proposed to mimic the learning of loss-splitting into multiple embeddings while significantly reducing the model size. In addition, we designed an improved solution to leverage additional metadata, such as camera ID and pose information, that uses 97% less parameters, further improving re-identification performance. In comparison to state-of-the-art (SoTA) methods, our approach outperforms competing solutions in Veri-776 by achieving 85.6% mAP and 97.7% CMC1 and obtains competitive results in Veri-Wild with 88.1% mAP and 96.3% CMC1. Overall, our work provides important insights into improving vehicle re-identification and presents a strong basis for other retrieval tasks. Our code is available at the link. Copyright © 2023, The Authors. All rights reserved.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

A data acquisition system to capture extreme human driving behaviour

A data acquisition system to capture extreme human driving b...

引用

IEEE International Conference on Autonomous Robot systems and Competitions (ICARSC)

作者： Ines Rito Lima José Santos-Victor Department of Electrical and Computer Engineering Instituto Superior Técnico Universidade de Lisboa Lisbon Portugal Institute for Systems and Robotics (ISR)) Instituto Superior Técnico Universidade de Lisboa Lisbon Portugal

Assessing and encoding human driving expertise while handling challenging events, primarily during drifts, can unveil new insights regarding the professional response to extreme driving which can be integrated into Autonomous Vehicles (AVs) safety procedures. Considering the complex and dynamic data related to both human behaviour and car devices, this research area urges the inclusion of Artificial Intelligence (AI) strategies for the extraction of interdependencies and patterns that encode human driving expertise. To do so, acquiring data that comprises the pilot and vehicle performances is of utmost importance. The present work proposes a specially designed data acquisition system to collect data regarding the vehicle dynamic movement and the associated human behaviour while driving in challenging environments. Distinct sensing devices are placed across the vehicle to retrieve the vehicle position, acceleration, angular rotation, steering orientation, among others. Data was collected from professional drivers during official racing events, and initial processing results show spatial and temporal coherence with the racing performances. The preliminary analysis of the data also reveals potentially extreme events during the race.

关键词： Performance evaluation Data acquisition Robot sensing systems Encoding Sensors Safety Data mining

来源：评论

学校读者我要写书评

暂无评论

The Role of the Essential Manifold in Data Mining – An Introductory Approach 23rd

The Role of the Essential Manifold in Data Mining – An I...

引用

23rd International Conference on Computational Science and Its Applications, ICCSA 2023

作者： de Fátima Pina, Maria Andrade, Marina Alexandra ISCTE - IUL University Institute of Lisbon Av. das Forças Armadas Lisbon1649-026 Portugal ISLA - Santarém Superior Institute of Management and Administration of Santarém Rua Teixeira Guedes 31 Santarém2000-029 Portugal ISR - UC Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Rua Silvio Lima - Pólo II Coimbra3030-290 Portugal ISTAR - IUL Information Sciences and Technologies and Architecture Research Center University Institute of Lisbon Av. das Forças Armadas Lisbon1649-026 Portugal

ISBN: (纸本)9783031371288

Interpolating data and the application of data mining techniques in nonlinear manifolds plays a significant role in different areas of knowledge, ranging from computer vision and robotics, to industrial and medical requests, and these growing number of applications have sparked the research interest of the scientific community to these topics. The Generalized Essential manifold, briefly, Essential manifold, consisting of the product of the Grassmann manifold of all k-dimensional subspaces of Rn and the Lie group of rotations in Rn, for instance, plays an important role in the problem of recovering the structure and motion from a sequence of images, also known as stereo matching, which is a crucial problem in image processing and computer vision. A well-known recursive procedure to generate interpolating polynomial curves in Euclidean spaces is the classical De Casteljau algorithm, which is a simple and powerful tool widely used in the field of Computer Aided Geometric Design, particularly because it is essentially geometrically based. This algorithm has been generalized to geodesically complete Riemannian manifolds. Thus, having this in mind, in this work we present all the ingredients for a detailed implementation of the generalized De Casteljau algorithm to generate geometric cubic polynomials in the Essential manifold preparing the ground to solve different real interpolation problems in this manifold. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

MonoVisual3DFilter: 3D tomatoes' localisation with monocular cameras using histogram filters

arXiv

引用

arXiv 2023年

作者： Magalhães, Sandro Costa dos Santos, Filipe Neves Moreira, António Paulo Dias, Jorge Faculty of Engineering University of Porto Porto Portugal INESC TEC Instituto de Engenharia de Sistemas e Computadores Tecnologia e Ciência Porto Portugal Institute of Systems and Robotics Department of Electrical Engineering and Computers University of Coimbra Coimbra Portugal Khalifa University of Science Technology and Research Abu Dhabi United Arab Emirates

Performing tasks in agriculture, such as fruit monitoring or harvesting, requires perceiving the objects' spatial position. RGB-D cameras are limited under open-field environments due to lightning interferences. So, in this study, we state to answer the research question: "How can we use and control monocular sensors to perceive objects' position in the 3D task space?" Towards this aim, we approached histogram filters (Bayesian discrete filters) to estimate the position of tomatoes in the tomato plant through the algorithm MonoVisual3DFilter. Two kernel filters were studied: the square kernel and the Gaussian kernel. The implemented algorithm was essayed in simulation, with and without Gaussian noise and random noise, and in a testbed at laboratory conditions. The algorithm reported a mean absolute error lower than 10 mm in simulation and 20 mm in the testbed at laboratory conditions with an assessing distance of about 0.5 m. So, the results are viable for real environments and should be improved at closer distances. © 2023, CC BY-NC-ND.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Edge computing: A neural network implementation on an iot device 5

Edge computing: A neural network implementation on an iot de...

引用

5th Experiment at International Conference, exp.at 2019

作者： Barreto, Ricardo Lobo, Jorge Menezes, Paulo ISR-Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal

ISBN: (纸本)9781728136370

This demonstration showcases the use of reconfigurable logic to implement edge computing for IoT devices able to provide specific information from raw data produced from some sensor, e.g. a camera or microphone, instead of the raw data itself. In what concerns the embedded processing capabilities, the focus is image processing using convolutional neuronal networks (CNN). This approach is clearly distinct from the current trends in IoT devices of using cloud computing to process the collected data. We intend a twist on the established paradigm and pursue an edge computing approach. Since we are targeting small and simple devices, we need some low power solution for the CNN computation. The demonstration will be made on a Terasic DE1 (SoC) reconfigurable system, with a field programmable gate array (FPGA), and hardwired ARM processor to build the IoT device. The collected data from the CNN computation, is transmitted using an IoT protocol to a broker. © 2019 IEEE.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Multispectral Vineyard Segmentation: A Deep Learning Comparison Study

arXiv

引用

arXiv 2021年

作者： Barros, Tiago Conde, P. Gonçalves, G. Premebida, C. Monteiro, M. Ferreira, C.S.S. Nunes, U.J. University of Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering Coimbra Portugal University of Coimbra Institute for Systems Engineering and Computers at Coimbra Department of Mathematics Coimbra Portugal Research Centre for Natural Resources Environment and Society Polytechnic Institute of Coimbra Coimbra Agrarian Technical School Coimbra Portugal Bolin Centre for Climate Research Department of Physical Geography Stockholm University Stockholm Sweden Navarino Environmental Observatory Navarino Dunes Messinia Greece

Digital agriculture has evolved significantly over the last few years due to the technological developments in automation and computational intelligence applied to the agricultural sector, including vineyards which are a relevant crop in the Mediterranean region. In this work, a study is presented of semantic segmentation for vine detection in real-world vineyards by exploring state-of-the-art deep segmentation networks and conventional unsupervised methods. Camera data have been collected on vineyards using an Unmanned Aerial System (UAS) equipped with a dual imaging sensor payload, namely a high-definition RGB camera and a five-band multispectral and thermal camera. Extensive experiments using deep-segmentation networks and unsupervised methods have been performed on multimodal datasets representing four distinct vineyards located in the central region of Portugal. The reported results indicate that SegNet, U-Net, and ModSegNet have equivalent overall performance in vine segmentation. The results also show that multimodality slightly improves the performance of vine segmentation, but the NIR spectrum alone generally is sufficient on most of the datasets. Furthermore, results suggest that high-definition RGB images produce equivalent or higher performance than any lower resolution multispectral band combination. Lastly, Deep Learning (DL) networks have higher overall performance than classical methods. The code and dataset are publicly available on https://***/Cybonic/***. © 2021, CC BY.

关键词： Precision agriculture

来源：评论

学校读者我要写书评

暂无评论

Adaptive Backstepping of Synergistic Hybrid Feedbacks with Application to Obstacle Avoidance

Adaptive Backstepping of Synergistic Hybrid Feedbacks with A...

引用

American Control Conference

作者： Pedro Casau Ricardo G. Sanfelice Carlos Silvestre Institute for Systems and Robotics (ISR) Instituto Superior Técnico Universidade de Lisboa Lisboa Portugal University of California Santa Cruz CA 95064 USA Department of Electrical and Computer Engineering of the Faculty of Science and Technology of the University of Macau Macau China

ISBN: (纸本)9781538679012;9781538679265

In this paper, we show that the hybrid controller that is induced by a Synergistic Lyapunov Function and Feedback (SLFF) pair relative to a compact set, can be extended to the case where the original affine control system is subject to a class of additive disturbances known as matched uncertainties, provided that the estimator dynamics do not add new equilibria to the closed-loop system. We also show that the proposed adaptive hybrid controller is amenable to backstepping. Finally, we apply the proposed hybrid control strategy to the problem of global asymptotic stabilization of a compact set in the presence of an obstacle and we illustrate this application by means of simulation results.

关键词： Obstacle avoidance Backstepping Feedback Collision avoidance Liapunov function control systems Closed loop systems Self tuning Controllers

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：