检索结果-内蒙古大学图书馆

Transformer-based multi-level attention integration network for video saliency prediction

Multimedia Tools and Applications 2025年第13期84卷 11833-11854页

作者： Tan, Rui Sun, Minghui Liang, Yanhua Software College Jilin University Qianjin Street Jilin Changchun130012 China College of Computer Science and Technology Jilin University Qianjin Street Jilin Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Qianjin Street Jilin Changchun130012 China

Most existing models for video saliency prediction heavily rely on 3D convolutional operations to extract spatio-temporal features. However, it is worth noting that 3D convolution produces a local receptive field, which may struggle to capture long-range spatio-temporal dependencies effectively. To compensate for such shortage, this paper introduces a novel approach called the Transformer-based Multi-level Attention Integration Network (TMAI-Net) for video saliency prediction. TMAI-Net is designed as a two-stream encoder-decoder model, carefully integrating multi-level features of semantic information. Our model incorporates a Multi-level Interactive Attention(MLIA) module and a Transformer, both implemented based on self-attention mechanism, which are placed at different levels of the model to capture long-range spatio-temporal feature dependencies. Additionally, our model operates on input video frames and attentional patches, allowing the Transformer module to capture structural similarities between related objects in global features and attention features. This, in turn, enables the model to allocate increased attention to salient areas. The efficacy of our proposed approach is validated through extensive experiments conducted on three widely recognized benchmark datasets. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Sparse Optical Flow Method for Camera Tracking under Fisheye Camera Models 5

A Sparse Optical Flow Method for Camera Tracking under Fishe...

引用

5th International Conference on Frontiers Technology of Information and Computer, ICFTIC 2023

作者： Zhang, Zuoming Wang, Xin College of Computer Science and Technology Jilin China College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin China

ISBN: (纸本)9798350309034

In computer image processing, optical flow is a classic task used to track the motion of pixels. In the field of robotics, methods like SLAM extensively employ sparse optical flow as a substitute for time-consuming feature point matching methods, enabling rapid and stable camera tracking. However, the widespread use of fisheye cameras introduces distortion issues in practical applications. Traditional sparse optical flow struggles to accurately track pixel positions in such scenarios, affecting the precision of camera pose tracking. Therefore, we introduce a sparse optical flow method designed for fisheye camera models, which correctly utilizes prior information obtained during camera tracking to enhance the accuracy of sparse optical flow. Finally, we propose a scheme for constructing a semi-distortion lookup table to accelerate distortion operations and reduce memory requirements. We conducted extensive testing on simulated datasets with various camera motion scenarios, and the experimental results demonstrate that our method outperforms OpenCV in terms of accuracy and versatility. © 2023 IEEE.

关键词： camera tracking component distortion fisheye camera model optical flow

来源：评论

学校读者我要写书评

暂无评论

A Graph Attribute Aggregation Method based on Feature engineering

引用

Journal of The Institution of Engineers (India): Series B 2022年第3期103卷 711-719页

作者： Wang, Hao Dong, Li-Yan Ma, Xin-Tao Sun, Ming-Hui College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

In the fields of social network analysis and knowledge graph, many semi-supervised learning algorithms based on graph convolutional neural network (GCN) have been widely used. Most of these algorithms usually improve the structure of the neural network and the sampling method of each layer of the neural network. However, they don’t pay much attention to the data pre-processing of the algorithm. In the analysis of the input data, the words of different quality in these original data are unevenly distributed. This may obscure some useful data and highlight some irrelevant data. In order to verify the correctness of this hypothesis, the paper proposes a feature matrix compression algorithm (FMC algorithm) for data pre-processing of GCN-based algorithms. The algorithm analyzes and arranges the word columns of the input matrix (the feature of graph) according to the frequency of the word, then merges those words in which the word frequency is smaller, so as to emphasize the role of these words in the graph and optimize the data scale. The present work uses four mainstream datasets in the field and several representative and different algorithms to complete the experiment. The experimental results show that the FMC algorithm achieves better performance. © 2021, The Institution of Engineers (India).

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Robust Federated Semi-Supervised Learning for Medical Image Classification via Pseudo-Label Filtering

Robust Federated Semi-Supervised Learning for Medical Image ...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Wang, Ziwei Guo, Shuyu Zhu, Mingzhu Bai, Tian Jilin University College of Software Changchun China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun China Jilin University College of Computer Science and Technology Changchun China

ISBN: (纸本)9798350386226

Federated learning (FL) enables collaborative model training across multiple medical institutions to ensure data security. However, due to the variations in medical imaging equipment and regions at different medical institutions, FL methods usually suffer from insufficient data annotations and irrelevant noise within private datasets. To address these issues, a robust federated semi-supervised learning method via pseudo-label filtering (PFRFed) is introduced to utilize unlabeled data while mitigating the impact of noise data. Compared with existing federated semi-supervised learning methods, we propose a pseudo-label filtering mechanism with double dynamic thresholds, which allows the model to adopt more unlabeled data by adjusting the confidence and entropy thresholds at each stage of model training. Moreover, to reduce the degradation caused by noise data in private datasets from different clients, a noise-tolerant loss function and a grouping aggregation method based on the local model similarity are employed. The comparative experiments demonstrate the effectiveness of PFRFed, which has achieved the best classification accuracy of 95.20% and 88.72% on two public medical datasets. Also, PFRFed exhibits heightened resilience to variations in noisy data ratio and labeled data ratio, reaffirming its versatility and robustness. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

TS-Net: Two-stage deformable medical image registration network based on new smooth constraints

引用

Magnetic Resonance Imaging 2023年第1期99卷 26-33页

作者： Song, Lei Ma, Mingrui Liu, Guixia College of Computer Science and Technology Jilin University Jilin Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Jilin Changchun 130012 China

Medical image registration can establish the spatial consistency of the corresponding anatomical structures between different medical images, which is important in medical image analysis. In recent years, with the rapid development of deep learning, the image registration methods based on deep learning greatly improve the speed, accuracy, and robustness of registration. Regrettably, these methods typically do not work well for large deformations and complex deformations in the image, and neglect to preserve the topological properties of the image during deformation. Aiming at these problems, we propose a new network TS-Net that learns deformation from coarse to fine and transmits information of different scales in the two stages. Two-stage network learning deformation from coarse to fine can gradually learn the large and complex deformations in images. In the second stage, the feature maps downsampled in the first stage for skip connection can expand the local receptive field and obtain more local information. The smooth constraints function used in the past is to impose the same restriction on the global, which is not targeted. In this paper, we propose a new smooth constraints function for each voxel deformation, which can better ensure the smoothness of the transformation and maintain the topological properties of the image. The experiments on brain datasets with complex deformations and heart datasets with large deformations show that our proposed method achieves better results while maintaining the topological properties of deformations compared to existing deep learning-based registration methods. © 2023 Elsevier Inc.

关键词： CNN Deformable image registration Smooth loss Two stage Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Migration: An Efficient Explorer Operator to Guide Swarm Evolution 1

Migration: An Efficient Explorer Operator to Guide Swarm Evo...

引用

1st IEEE International Conference on Medical Artificial Intelligence, MedAI 2023

作者： Han, Xiaosong Zhou, Chengyang Liang, Yanchun Guan, Renchu College of Computer Science and Technology Jilin University Key Laboratory for Symbol Computation and Knowledge Engineering National Education Ministry Changchun China College of Computer Science and Technology Zhuhai Laboratory of Key Laboratory for Symbol Computation and Knowledge Engineering Ministry of Education Jilin University Zhuhai College of Science and Technology Key Laboratory for Symbol Computation and Knowledge Engineering National Education Ministry Zhuhai China

ISBN: (纸本)9798350358780

Particle Swarm Optimization with Migration (MPSO) is proposed to solve the issue that PSO will encounter unbearable time cost problems when dealing with High-dimension, Expensive and Black-box objective function tasks. The Migration operator is inspired by the migration of Salmon. Salmon will start a dangerous journey from the ocean to the home rivers for reproduction. The process of the entire behavior is similar to the reduction and recovery of dimension. Therefore, we design the Migration operator where a pretrained Wasserstein Autoencoders (WAE) is applied to simulate the migration behavior to accelerate the process of evolution in PSO, and we use Least-Squares Regression in lower space to produce better generation. In comparison with famous baseline methods in some benchmark functions, MPSO converges faster and more accurately, which shows the great potential of migration operations. © 2023 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

An embedded vertical‐federated feature selection algorithm based on particle swarm optimisation

引用

CAAI Transactions on Intelligence Technology 2023年第3期8卷 734-754页

作者： Yong Zhang Ying Hu Xiaozhi Gao Dunwei Gong Yinan Guo Kaizhou Gao Wanqiu Zhang School of Information and Control Engineering China University of Mining and TechnologyXuzhouChina The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin UniversityChangchunChina School of Computing University of Eastern FinlandKuopioFinland The Macao Institute of Systems Engineering Macao University of Science and TechnologyTaipaChina

In real life,a large amount of data describing the same learning task may be stored in different institutions(called participants),and these data cannot be shared among par-ticipants due to privacy *** case that different attributes/features of the same instance are stored in different institutions is called vertically distributed *** pur-pose of vertical‐federated feature selection(FS)is to reduce the feature dimension of vertical distributed data jointly without sharing local original data so that the feature subset obtained has the same or better performance as the original feature *** solve this problem,in the paper,an embedded vertical‐federated FS algorithm based on particle swarm optimisation(PSO‐EVFFS)is proposed by incorporating evolutionary FS into the SecureBoost framework for the first *** optimising both hyper‐parameters of the XGBoost model and feature subsets,PSO‐EVFFS can obtain a feature subset,which makes the XGBoost model more *** the same time,since different participants only share insensitive parameters such as model loss function,PSO‐EVFFS can effec-tively ensure the privacy of participants'***,an ensemble ranking strategy of feature importance based on the XGBoost tree model is developed to effectively remove irrelevant features on each ***,the proposed algorithm is applied to 10 test datasets and compared with three typical vertical‐federated learning frameworks and two variants of the proposed algorithm with different initialisation ***-mental results show that the proposed algorithm can significantly improve the classifi-cation performance of selected feature subsets while fully protecting the data privacy of all participants.

关键词： Evolutionary optimization feature selection privacy protection vertical‐federated learning

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning Path Planning based on Step Batch Q-Learning Algorithm

Reinforcement Learning Path Planning based on Step Batch Q-L...

引用

2022 IEEE International Conference on Artificial Intelligence and Computer Applications, ICAICA 2022

作者： Yin, Zhiqian Cao, Wei Song, Tao Yang, Xu Zhang, Tianhao Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education College of Software Changchun130012 China

ISBN: (数字)9781665499910

ISBN: (纸本)9781665499910

With the rapid development of artificial intelligence, people have put forward higher requirements for robot path planning. As a more commonly used algorithm, reinforcement learning learns from experience by imitating the process of human learning skills and continuously iterates to do. In this paper, an improved Step Batch Q-Learning algorithm is proposed to solve the problem that the traditional Q-Learning algorithm has a slow convergence speed in discrete states. Prediction can detect obstacles and target points earlier, reduce the number of iterations, and significantly speed up the convergence time. Under the same experimental conditions, the method in this paper is compared with the traditional Q-Learning algorithm. The experimental results show that the improved Step Batch Q-Learning algorithm has significantly improved convergence speed and good adaptability. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

BSDIFF Difference Algorithm Based on LZMA2 for In-Vehicle ECUs 7th

BSDIFF Difference Algorithm Based on LZMA2 for In-Vehicle EC...

引用

7th International Conference on Computing, Control and Industrial engineering, CCIE 2023

作者： Li, Zhihao Qin, Guihe Liang, Yanhua Danzengouzhu Yang, Lifeng College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

ISBN: (纸本)9789819927296

With cars becoming the primary choice of people’s daily travel, on-board OTA upgrade technology has arisen a common concern in the automotive industry. Traditional on-board OTA upgrade technology, which adopts a method of updating the whole package, successfully avoids offline upgrade. However, it takes too long to download the whole update package and upgrade in-vehicle ECUs because of the huge data transmission. In addition, packet loss and disconnection are easy to occur in case of network congestion. As a result, it will seriously affect users’ online upgrade experience. Therefore, incremental update technology should be used instead of updating the whole package in on-board OTA upgrade technology. In order to solve above problems, in this paper, we propose an improved BSDIFF difference algorithm based on LZMA2 compression algorithm to realize the incremental update of on-board OTA. Traditional BSDIFF can generate a patch file according to an old app and a new app. And a patch file is far smaller than the corresponding new app, which is conducive to remote transmission and reduces the download time of users. Through our experiments, LZMA2-based BSDIFF has a better compression ratio than traditional BSDIFF, which is more suitable for the incremental update of in-vehicle ECUs with limited memory resources. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Automotive industry

来源：评论

学校读者我要写书评

暂无评论

Discovering Skyline Periodic Itemset Patterns in Transaction Sequences 9th

Discovering Skyline Periodic Itemset Patterns in Transactio...

引用

19th International Conference on Advanced Data Mining and Applications, ADMA 2023

作者： Chen, Guisheng Li, Zhanshan College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

ISBN: (纸本)9783031466601

As an extended version of frequent itemset patterns, periodic itemset patterns concern both the frequency and periodicity of itemsets at the same time, so they contain more information than frequent itemset patterns, which only concern the frequency. With further research, we found that, in some cases, the periodic itemset patterns with higher frequency, or with optimal periodicity, or with both higher frequency and optimal periodicity have higher application value. However, there is currently no work focusing on such a kind of periodic itemset patterns. In view of this, this paper first proposes a new concept of skyline periodic itemset patterns, and states the problem of skyline periodic itemset pattern mining, then presents an algorithm called SLPIM (SkyLine Periodic Itemset pattern Miner) for skyline periodic itemset pattern mining. SLPIM first adopts the well-known FP-Growth algorithm to mine all frequent itemset patterns, and then uses an effective judgment strategy to determine which frequent itemset patterns are skyline periodic itemset patterns. Finally, experiments are conducted on two real-world and two simulated datasets. The results show that SLPIM is competent for mining skyline periodic itemset patterns. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Data mining

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：