检索结果-内蒙古大学图书馆

GPS: graph contrastive learning via multi-scale augmented views from adversarial pooling

science China(information sciences) 2025年第1期68卷 145-158页

作者： Wei JU Yiyang GU Zhengyang MAO Ziyue QIAO Yifang QIN Xiao LUO Hui XIONG Ming ZHANG School of Computer Science National Key Laboratory for Multimedia Information ProcessingPeking University Artificial Intelligence Thrust The Hong Kong University of Science and Technology Department of Computer Science University of California

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views(i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast(GPS) to address these *** by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关键词： graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling

来源：评论

学校读者我要写书评

暂无评论

Complementary Dual-Branch Network for Space-Time Video Super-Resolution 27th

Complementary Dual-Branch Network for Space-Time Video Supe...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Tian, Ming Li, Tianyi Luo, RongSheng Gao, Changxin Sang, Nong National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9783031781247

Space-time video super-resolution aims to simultaneously increase the space-time resolution of low-resolution and low frame-rate videos. Existing deep learning-based methods have made notable strides, predominantly achieving space-time video super-resolution through the relatively simple integration of modules for video super-resolution and video frame interpolation sub-tasks. However, these methods typically do not fully exploit the inherent relationships between the two sub-tasks. To address this limitation, we propose a Complementary Dual-Branch Network designed to better explore the interdependence of the two sub-tasks. Specifically, our dual-branch architecture facilitates mutual enhancement between video super-resolution and video frame interpolation sub-tasks within each branch and provides mutual guidance between the two branches. Additionally, we introduce a simple yet effective strategy for the rough estimation of optical flow, incorporating Flow-Guided Deformable Alignment into space-time video super-resolution to achieve precise motion estimation. In addition, we use an RNN-based Backward and Forward Recurrent module to ensure that all frames can utilize the information of the whole sequence. It is more efficient and memory saving compared to the currently popular bidirectional LSTM module. Experimental results on several datasets show that our method achieves superior accuracy and requires fewer parameters compared to state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Neighborhood Structure Consistency for Point Cloud Registration With Severe Outliers

引用

IEEE Sensors Journal 2025年第11期25卷 20209-20223页

作者： Ding, Junfeng An, Pei Xu, Yunbiao Ma, Jie Huazhong University of Science and Technology National Key Laboratory of Science and Technology on Multispectral Information Processing School of Artificial Intelligence and Automation Wuhan430074 China Huazhong University of Science and Technology School of Electronic Information and Communication Wuhan430074 China

Point cloud registration is a fundamental task in 3D sensor technology and computer vision, aiming to estimate the rigid pose between two point clouds and align them. Correspondence-based methods generate and verify hypothetical poses by sampling consistent correspondences. However, the presence of severe outliers in feature correspondences poses substantial challenges during the sampling process. It may result in misjudgment of inlier consistency relationships, leading to incorrect sampling and causing point cloud registration failure. In this paper, we introduce the Neighborhood Structure Consistency (NSC) metric to guide consensus excavation. It incorporates tighter global consistency constraints rather than local consistency, yielding a more distinct inlier cluster and significantly mitigating the impact of outliers. After that, we propose a Neighborhood Structure Consistency based-Graph guided Sampling Consensus (NSC-GSAC) algorithm, which performs the sampling process on a consistency graph based on the NSC metric. This graph provides a dense representation of connections among inlier correspondences, narrowing the search space and facilitating cleaner consensus subsets sampling, thus improving the registration robustness and efficiency. Additionally, to further improve efficiency, we introduce a Pruning Redundancy Hypothesis Selection (PRHS) strategy to prioritize highly promising representative hypothetical poses, thus accelerating the registration process. Extensive experiments on real-world indoor and outdoor datasets demonstrate that our method achieves superior registration accuracy and efficiency compared to state-of-the-art methods, including both traditional and learning-based methods. Furthermore, experimental results also indicate that our method exhibits more robust performance in the presence of severe outliers, outperforming all other methods. © 2001-2012 IEEE.

关键词： Three dimensional computer graphics

来源：评论

学校读者我要写书评

暂无评论

Human pose estimation constrained by human structure information 12

Human pose estimation constrained by human structure informa...

引用

SPIE 12th International Symposium on multispectral Image processing and Pattern Recognition, MIPPR 2023

作者： Zou, Lamei Wang, Hao Xie, Jia Wu, Chengqing Lu, Luhan Guo, Yichun National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9781510674967

ISBN: (纸本)9781510674950

We present a two-dimensional human pose estimation network constrained by the human structure information (HSINet). HSINet effectively fuses features of different scales and explicitly integrates human structure information to enhance the precision of key point localization. The architecture of HSINet comprises three pivotal modules: The feature extraction module, the encoding module, and the decoding module. The feature extraction module within HSINet employs the architecture of High-Resolution Net (HRNet). In contrast to HRNet, we remove redundant layers, and enhance the ability to combine global features and local features using the Gated Attention Unit (GAU). The second module encodes the feature maps derived from the feature extraction module. Each feature map corresponds to a joint point and is characterized by two feature vectors representing the x and y axes. Utilizing graph convolution for encoding introduces constraints based on human structure information. Subsequently, these encoded feature maps are decoded into precise coordinates of key points. The experiment results on COCO datasets show that our proposed method can improve the precision of key point detection while effectively reducing the number of parameters. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

引用

2024 International Conference on Image, Signal processing, and Pattern Recognition, ISPP 2024

作者： Yu, Pan Zhao, Wei Huang, Yan Wang, Guoyou National Key Laboratory of Science and Technology on Multispectral Information Processing School of Artificial Intelligence and Automation Huazhong University of Science and Technology Hubei Wuhan430074 China

ISBN: (纸本)9781510680425

Infrared images are widely used in security monitoring and autonomous driving due to their resistance to light changes and adverse weather conditions. But, the low contrast and colorless characteristics of infrared images limit the effectiveness of human observation and subsequent detection and recognition algorithms. Methods for translating infrared images into visible images can overcome the above shortcomings, among which contrastive learning methods using self-similarity features achieve the best performance. However, methods based on contrastive learning face the sub-optimization problem caused by random sampling. And the contrastive loss based on self-similarity features faces the problem of encoding entanglement when used for infrared-visible image translation, that is, the features extracted from different categories of regions cannot be distinguished. Therefore, we propose a cross-similarity guided contrastive learning method for infrared-visible image translation. First, to address the randomness and inefficiency of the contrastive loss random sampling process, a sampling strategy based on information entropy ranking of cross-similarity matrix is proposed to obtain sampling points for subsequent contrastive loss calculation. By calculating the information entropy of cross-similarity matrix between input and generated images and sorting them, the sampling points with the most information can be obtained. Second, to alleviate the encoding entanglement problem of the self-similarity contrastive loss due to the low contrast of infrared images, multi-scale spatially adjacent graph structure consistency loss and spatially separated graph structure consistency loss based on cross-similarity matrices are proposed. Experiments on KAIST and FLIR datasets show that the proposed method has the best score and visual performance compared with multiple advanced infrared-visible translation methods. Ablation experiments further illustrate the effectiveness of the method.

关键词： Infrared imaging

来源：评论

学校读者我要写书评

暂无评论

A shape and texture feature blending network for bone marrow cells classification using auxiliary learning 12

A shape and texture feature blending network for bone marrow...

引用

SPIE 12th International Symposium on multispectral Image processing and Pattern Recognition, MIPPR 2023

作者： Guo, Zhengyang Yang, Zengbiao Xiang, Xuanyu Tan, Yihua National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

ISBN: (数字)9781510674981

ISBN: (纸本)9781510674974

Morphological examination of bone marrow cells is crucial for diagnosing blood diseases. However, manual classification of bone marrow cells is time-consuming and subjective. Therefore, it is necessary to develop an autoclassification method for bone marrow cells. Although deep learning methods are commonly used for cell classification like Resnet50, they don't take advantage of the features of bone marrow cells such as the shape features of cells. However, the shape of cell and nucleus plays a significant role in distinguishing between different cell types. In this paper, we proposed a Shape and Texture Feature Blending Network (STFB-Net) for bone marrow cells classification based on auxiliary learning. We used ResNext50 as the backbone network for STFB-Net due to its exceptional ability to extract texture features from cells. In addition, we proposed the Shape Feature Extraction Module (SFEM) to enhance the backbone network's ability for capturing shape features. SFEM shares a part of parameters with the backbone network. SFEM extracts features at multiple scales and up-samples them, then fuses multi-scale features to predict the shape of cells. We performed experiments on two bone marrow cell datasets. The results show that the proposed STFB-Net can effectively extract texture and shape features, which brings better performance than other cell classification methods. By using the Grad-CAM method to visualize the features extracted by STFB-Net, we proved the reliability and the effectiveness of the STFB-Net in extracting cell shape features. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Time-varying Formation-containment Tracking Control of Heterogeneous Multi-agent System under Denial of Service Attack and Sensor Fault

Time-varying Formation-containment Tracking Control of Heter...

引用

2024 China Automation Congress, CAC 2024

作者： Niu, Baojian Fan, Huijin Liu, Lei Wang, Bo National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Hubei Wuhan China

ISBN: (纸本)9798350368604

In this paper, the problem of output-feedback time-varying formation-containment tracking control for heterogeneous multi-agent systems with sensor fault under distributed denial of service (DoS) attacks is studied. The leaders are non-cooperative, which are embedded with unknown input. Firstly, in order to estimate the sensor fault and the states of the followers, a distributed fault-detecting observer is designed. Secondly, in order to deal with distributed DoS attacks and the unknown input of the leaders, a leader state estimator is designed. Finally, a distributed output-feedback containment control law is proposed to realize the time-varying formation tracking of the convex hull of multiple leaders. It is proved that the proposed control scheme guarantees the exponential convergence of all errors of the multi-agent system, and the effectiveness and superiority of the proposed scheme are verified by simulation experiments. © 2024 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Relative Position-based Distributed Fixed-time Formation Control of Multiple UAVs under GNSS-denied environment 43

Relative Position-based Distributed Fixed-time Formation Con...

引用

43rd Chinese Control Conference, CCC 2024

作者： Yang, Jiahao Zhi, Yongran Xue, Chenlong Liu, Lei Fan, Huijin Wang, Bo Huazhong University of Science and Technology National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Wuhan430074 China

ISBN: (纸本)9789887581581

In this paper, a novel fixed-time formation control algorithm is proposed for multiple unmanned aerial vehicles (UAVs) in environments where global navigation satellite system(GNSS) is denied, where the absolute positions of UAVs are unmeasurable, and only relative positions are accessible. To address this challenge, firstly, a composite observers is designed with one estimating relative positions between leader and followers and the other being a distributed fixed-time velocity observer. The integration of these observers enables each follower to obtain relative position and velocity of the leader within a fixed time. Subsequently, based on the composite observers and the desired formation matrix, a fixed-time controller is formulated aiming to achieve the formation task within a fixed time. Furthermore, Lyapunov stability theory demonstrates that the designed observers and controller can successfully accomplish the formation task within a fixed time. Finally, numerical simulations validate the effectiveness of the analytical results. © 2024 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Adaptive Tracking of Uncertain Markovian Jump Nonlinear Systems With Random Actuator Faults

Adaptive Tracking of Uncertain Markovian Jump Nonlinear Syst...

引用

2024 China Automation Congress, CAC 2024

作者： Zhang, Jiao-Yang Fan, Huijin Liu, Lei Wang, Bo National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9798350368604

This article deals with the adaptive fault-tolerant tracking problem for uncertain Markovian jump nonlinear systems (MJNSs) in a stochastic framework. A joint and a scalar Markovian process are employed to describe behaviors of the unknown parameters and actuator faults, respectively, thereby characterizing uncertainties in a random manner. With the aid of the backstepping technique, an adaptive tracking controller as well as the corresponding parameter update laws is constructed by treating the regimes of all Markovian variables as constants. It is shown that with appropriate design parameters, the output tracking error can converge into a small adjustable residual set in the sense of mean quadratic value while all closed-loop signals are bounded in probability, regardless of infinite number of random abrupt jumps in both uncertain parameters and the control input. An illustrative example is provided to show the effectiveness of the theoretical results. © 2024 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

An Edge-Guided Network for Building Footprint Extraction from Optical Remote Sensing Images 12

An Edge-Guided Network for Building Footprint Extraction fro...

引用

SPIE 12th International Symposium on multispectral Image processing and Pattern Recognition, MIPPR 2023

作者： Zhou, Yuhang Tan, Yihua National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

Accurate building footprint extraction from optical remote sensing images remains challenging due to the diverse appearance and complex scenarios. Although recent deep learning-based methods have been shown to greatly improve the accuracy of building footprint extraction, vanilla deep networks still suffer from ambiguous predictions of edge pixels. The building edge contains abundant location and shape information, which is important for downstream applications such as building positioning and area measurement. Therefore, the problem of inaccurate edge prediction needs to be resolved urgently. To this end, we propose a novel edge-guided network (EGNet) that makes ample use of the edge prior in an end-to-end manner. First, an edge extraction module (EEM) is proposed to extract the building edge map. Then, an edge-guidance module (EGM) is designed to utilize the edge map to guide each encoder block in extracting edge-related features. Furthermore, a multi-scale context aggregation module (MCAM) is built to enhance the feature representation by aggregating contextual semantics with different receptive fields. EGNet can effectively mine edge semantics and guide the representation learning of boundaries, achieving 75.21% and 91.16% IoU on the Massachusetts and WHU datasets, respectively. Experimental results demonstrate that the proposed EGNet has a certain superiority in both accuracy and efficiency compared with current state-of-the-art (SOTA) methods. © 2024 SPIE.

关键词： Extraction

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：