检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Shi, Xinyu Wei, Dong Zhang, Yu Lu, Donghuan Ning, Munan Chen, Jiashun Ma, Kai Zheng, Yefeng School of Computer Science & Engineering Key Lab of Computer Network & Information Integration Ministry of Education Southeast Univ. Nanjing China Tencent Jarvis Lab Shenzhen China

Research into Few-shot Semantic Segmentation (FSS) has attracted great attention, with the goal to segment target objects in a query image given only a few annotated support images of the target class. A key to this challenging task is to fully utilize the information in the support images by exploiting fine-grained correlations between the query and support images. However, most existing approaches either compressed the support information into a few class-wise prototypes, or used partial support information (e.g., only foreground) at the pixel level, causing non-negligible information loss. In this paper, we propose Dense pixel-wise Cross-query-and-support Attention weighted Mask Aggregation (DCAMA), where both foreground and background support information are fully exploited via multi-level pixel-wise correlations between paired query and support features. Implemented with the scaled dot-product attention in the Transformer architecture, DCAMA treats every query pixel as a token, computes its similarities with all support pixels, and predicts its segmentation label as an additive aggregation of all the support pixels’ labels—weighted by the similarities. Based on the unique formulation of DCAMA, we further propose efficient and effective one-pass inference for n-shot segmentation, where pixels of all support images are collected for the mask aggregation at once. Experiments show that our DCAMA significantly advances the state of the art on standard FSS benchmarks of PASCAL-5i, COCO-20i, and FSS-1000, e.g., with 3.1%, 9.7%, and 3.6% absolute improvements in 1-shot mIoU over previous best records. Ablative studies also verify the design DCAMA. © 2022, CC BY-NC-ND.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Medical Image Segmentation Use Convolutional Attention Augmentation TransUNet with Skip Connection Enhancement

Medical Image Segmentation Use Convolutional Attention Augme...

引用

International Conference on computer Supported Cooperative Work in Design

作者： Qingzhi Zou Jing Zhao Ming Li Ling Chen Lin Yuan Ronghuan Zhang Yushuai Hu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China School of Intelligence and Information Engineering Shandong University of Traditional Chinese Medicine Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

The primary objective of medical image segmentation is the isolation of pathological tissues and distinct organs from medical images, thereby assisting in medical diagnosis. The methods employed for medical image segmentation encompass convolutional neural networks and transformer-based methods. Although the self-attention mechanism in Transformers improves its capability to capture long-range dependencies, it have limitations in learning local (contextual) relationships between pixels. Some previous research has tried to solve this problem by incorporating convolutional layers into the encoder or decoder of transformers, but sometimes feature inconsistencies arise. To better extract local features from images using convolutional neural networks and to fuse low-resolution and high-resolution features from higher-level and lower features, we propose the Convolutional Attention Augmentation TransUNet (CAA-TransUNet) model. In our model, firstly, we propose a convolutional attention augmentation module that enhances both local and global features by suppressing irrelevant background information. Secondly, we have integrated attention gates into the skip connections to aggregate feature information from various stages of the encoder during the upsampling process. Finally, we employ the technique of aggregating the loss of multi-stage features to expedite convergence speed and enhance performance. The experimental results on three public datasets demonstrate that our proposed model significantly outperforms the baseline methods.

关键词： Image segmentation Pathology Computational modeling Aggregates Logic gates Feature extraction Transformers

来源：评论

学校读者我要写书评

暂无评论

Estimator Meets Equilibrium Perspective: A Rectified Straight Through Estimator for Binary Neural networks Training

arXiv

引用

arXiv 2023年

作者： Wu, Xiao-Ming Zheng, Dian Liu, Zuhao Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Pengcheng Lab China Guangdong Province Key Laboratory of Information Security Technology China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

Binarization of neural networks is a dominant paradigm in neural networks compression. The pioneering work BinaryConnect uses Straight Through Estimator (STE) to mimic the gradients of the sign function, but it also causes the crucial inconsistency problem. Most of the previous methods design different estimators instead of STE to mitigate it. However, they ignore the fact that when reducing the estimating error, the gradient stability will decrease concomitantly. These highly divergent gradients will harm the model training and increase the risk of gradient vanishing and gradient exploding. To fully take the gradient stability into consideration, we present a new perspective to the BNNs training, regarding it as the equilibrium between the estimating error and the gradient stability. In this view, we firstly design two indicators to quantitatively demonstrate the equilibrium phenomenon. In addition, in order to balance the estimating error and the gradient stability well, we revise the original straight through estimator and propose a power function based estimator, Rectified Straight Through Estimator (ReSTE for short). Comparing to other estimators, ReSTE is rational and capable of flexibly balancing the estimating error with the gradient stability. Extensive experiments on CIFAR-10 and ImageNet datasets show that ReSTE has excellent performance and surpasses the state-of-the-art methods without any auxiliary modules or losses. Copyright © 2023, The Authors. All rights reserved.

关键词： Stability

来源：评论

学校读者我要写书评

暂无评论

A Cross-Modal Interactive Memory network Based on Fine-Grained Medical Feature Extraction for Radiology Report Generation

A Cross-Modal Interactive Memory Network Based on Fine-Grain...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Xitong Ma Yuansen Kuang Lin Yuan Cheng Tian Yijie Zeng Song Liu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9781665410205

ISBN: (纸本)9781665410212

Radiology report generation is an essential task in the medical field, which aims to automate the generation of medical terminology descriptions of radiology images. However, this task currently suffers from several problems: 1) existing methods need to manually build knowledge graphs or templates (consuming time and effort) to introduce medical or prior knowledge to assist in report generation; 2) previous models cannot handle the problem of data bias well (anomaly reports and anomaly descriptions make up only a tiny portion of the dataset), causing the models to ignore the learning of anomaly descriptions easily; 3) existing approaches cannot robustly supervise the model, resulting in incomplete and inconsistent reports being generated. To address these issues, we propose a cross-modal interactive memory network based on fine-grained medical feature extraction. In our model, we design a cross-modal interactive memory network to automatically store and remember the required medical text knowledge and use this medical knowledge to help generate reports. Furthermore, we design an abnormal medical knowledge enhancement module to enhance the learning of abnormal fine-grained knowledge through the interaction of disease topics and their states to interact with text features. In addition, we design a cross-modal joint semantic loss unit to reduce semantic differences between different features and improve the visual representation ability of the model. We experimented and evaluated our model on MIMIC-CXR and IU-Xray datasets to compare with other baseline models.

关键词： Knowledge engineering Visualization Terminology Semantics MIMICs Memory management Radiology Feature extraction Data models Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

ERC DMSP: Emotion Recognition in Conversation Based on Dynamic Modeling of Speaker Personalities

ERC DMSP: Emotion Recognition in Conversation Based on Dynam...

引用

International Joint Conference on Neural networks (IJCNN)

作者： Xiaoyang Li Zhenyu Yang Zhijun Li Yiwen Li Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Emotion Recognition in Conversations (ERC) is a widely applicable task, whether in emotional chatbots or recommender systems for dialog scenarios. The ability of systems to accurately identify human emotions is a crucial component in ERC success. In everyday conversations, a speaker’s emotional expression is closely tied to their individual characteristics and personality traits. Furthermore, individuals exhibit significant variations in their expressions of the same emotion. Traditional methods for ERC often overlook these personality differences and dynamic changes, leading to unreliable results. To address this issue, we propose a conversation emotion recognition model based on dynamic modeling of speaker’s personality (ERC DMSP). In our approach, to extract a speaker’s personality profile, we design a personality capture module to model the personalities of individuals based on their historical utterances. Moreover, a bidirectional influence exists between a speaker’s personality tendencies and their linguistic behavior. The speaker’s personality profile is thus continually updated based on the ongoing dialog. Additionally, since the personalities of other speakers in the conversation can also affect the form and intensity of a speaker’s emotional expression, we construct a conversation emotion representation utilizing group personality traits. We then comprehensively incorporate the speaker’s personality into the emotion representation of each utterance at both the group and individual levels through the personality engagement module. Throughout this process, the representation of dialog and the speaker’s personality undergo hierarchical updates, leading to more accurate conversation emotion recognition. We conducted extensive experiments on four commonly used public benchmark datasets for ERC to evaluate the effectiveness of our proposed ERC DMSP model. The results demonstrate its efficacy, as it exhibits a significant improvement over other methods in popular

关键词： Measurement Emotion recognition Accuracy Codes Neural networks Oral communication Color

来源：评论

学校读者我要写书评

暂无评论

Minimization of Age of information for Satellite-Terrestrial Covert Communication with a Full-Duplex Receiver

Minimization of Age of Information for Satellite-Terrestrial...

引用

International Conference on networking and network Applications (NaNA)

作者： Yizhi Cao Zewei Guo Qifeng Miao Ranran Sun Ji He Xiaochen Li Hangzhou Institute of Technology Xidian University Hangzhou China Shaanxi Key Laboratory of Network and System Security Xidian University Xi’an China School of Systems Information Science Future University Hakodate Hakodate Japan School of Computer Science and Technology Xidian University Xi’an China PLA Strategic Support Force Information Engineering University Zhengzhou China

ISBN: (数字)9798350376777

ISBN: (纸本)9798350376784

This paper investigates the freshness of information for covert communication in a satellite-terrestrial system consisting of a satellite transmitter Alice, a terrestrial full-duplex receiver Bob, and a terrestrial warden Willie. Particularly, Alice tries to transmit covert information to Bob under the detection of Willie, while Bob emits jamming signals to confuse Willie simultaneously. We first provide theoretical modeling for the average age of information (AOI) of Bob and the covert constraint of Willie. Based on the theoretical model, we then drive a optimal transmit power of Alice to minimize the average AoI with the constraint of covertness requirement. Finally, numerical results are presented to verify the theoretical results and to illustrate the impacts of transmit and jamming power on the covertness and average AoI.

关键词： Satellites Transmitters Rician channels Receivers Full-duplex system information age Space-air-ground integrated networks

来源：评论

学校读者我要写书评

暂无评论

Dynamic Pseudonym Semantic-location Privacy Protection Based on Continuous Query for Road network

引用

International Journal of network security 2021年第4期23卷 642-649页

作者： Wang, Yonglu Zuo, Kaizhong Liu, Rui Zhao, Jun School of Computer and Information Anhui Normal University Anhui Provincial Key Lab of Network and Information Security Anhui Normal University Wuhu241002 China

Continuous queries are one of the most common queries in location-based services, although particularly useful, such queries raise serious privacy concerns. Due to frequently updated continuous locations, continuous query privacy will be disclosed. Constructing a constant anonymous user set is one of the main methods used to protect privacy information. The exiting methods don’t consider query duration or users’ personalized privacy requirements, and the semantic information of location isn’t fully taken into account. In order to solve these problems, a dynamic pseudonym semantic-location privacy protection algorithm based on continuous query for road network is proposed in this paper. Firstly, the city road network in question is divided using the Voronoi diagram. Secondly, a secure anonymous region is generated by adopting the greedy strategy based on location and privacy requirement. Finally, the intersection of anonymous user sets at different times is calculated, and the user’s identity information is updated using the dynamic pseudonym mechanism. Experimental results show that the proposed method can satisfy users’ personalized privacy requirements and improve the security of privacy protection and query efficiency. © 2021

关键词： Location based services

来源：评论

学校读者我要写书评

暂无评论

ASAG: Building Strong One-Decoder-Layer Sparse Detectors via Adaptive Sparse Anchor Generation

ASAG: Building Strong One-Decoder-Layer Sparse Detectors via...

引用

International Conference on computer Vision (ICCV)

作者： Shenghao Fu Junkai Yan Yipeng Gao Xiaohua Xie Wei-Shi Zheng School of Computer Science and Engineering Sun Yat-sen University China Guangdong Province Key Laboratory of Information Security Technology China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Pengcheng Lab China

Recent sparse detectors with multiple, e.g. six, decoder layers achieve promising performance but much inference time due to complex heads. Previous works have explored using dense priors as initialization and built one-decoder-layer detectors. Although they gain remarkable acceleration, their performance still lags behind their six-decoder-layer counterparts by a large margin. In this work, we aim to bridge this performance gap while retaining fast speed. We find that the architecture discrepancy between dense and sparse detectors leads to feature conflict, hampering the performance of one-decoder-layer detectors. Thus we propose Adaptive Sparse Anchor Generator (ASAG) which predicts dynamic anchors on patches rather than grids in a sparse way so that it alleviates the feature conflict problem. For each image, ASAG dynamically selects which feature maps and which locations to predict, forming a fully adaptive way to generate image-specific anchors. Further, a simple and effective Query Weighting method eases the training instability from adaptiveness. Extensive experiments show that our method outperforms dense-initialized ones and achieves a better speed-accuracy trade-off. The code is available at https://***/iSEE-laboratory/ASAG.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LGDB-Net: Dual-Branch Path for Building Extraction from Remote Sensing Image

LGDB-Net: Dual-Branch Path for Building Extraction from Remo...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Ronghuan Zhang Jing Zhao Ming Li Qingzhi Zou Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China School of Intelligence and Information Engineering Shandong University of Traditional Chinese Medicine Jinan China

ISBN: (数字)9798331515966

ISBN: (纸本)9798331515973

Extracting buildings from remote sensing images using deep learning techniques is a widely applied and crucial task. Convolutional Neural networks (CNNs) adopt hierarchical feature representation, showcasing powerful capabilities in extracting local information but facing challenges in capturing global features. Transformers can address this limitation, but they perform poorly in extracting local features and significantly increase memory requirements and computational complexity. To overcome these challenges, we propose a method for building extraction from remote sensing images called LGDB-Net (Local-Global Dual-Branch network), employing a dual-branch approach. Firstly, inspired by Swin Transformer, we designed GB-Former(Global Branch-Former) as the backbone network to model global information. We use a linear multi-head self-attention mechanism to reduce time and memory complexity while maintaining a large global receptive field. Additionally, we replace the traditional multi-layer perceptron with a convolution-enhanced multi-layer perceptron to improve channel feature representation, reduce model parameters, and enhance segmentation performance. Secondly, we use multiple Depth-wise Conv3×3 + LN (Layer Normalization) + GeLU (Gaussian Error Linear Unit) modules as the auxiliary branch for local detailed feature extraction. Finally, we adopt a multi-scale feature fusion strategy to integrate feature information from both branches. We conduct a series of experiments on three datasets: WHU, Massachusetts, and Inria. The experimental results demonstrate that the proposed method not only effectively improves segmentation accuracy with lower building omission and commission rates but also significantly reduces model parameters and computational complexity.

关键词： Accuracy Computational modeling Memory management Neural networks Feature extraction Transformers Sensors Data mining Computational complexity Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised 3D Scene Flow Estimation Guided by Superpoints

Self-Supervised 3D Scene Flow Estimation Guided by Superpoin...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Yaqi Shen Le Hui Jin Xie Jian Yang PCA Lab Nanjing University of Science and Technology Nanjing China PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

3D scene flow estimation aims to estimate point-wise motions between two consecutive frames of point clouds. Superpoints, i.e., points with similar geometric features, are usually employed to capture similar motions of local regions in 3D scenes for scene flow estimation. However, in existing methods, superpoints are generated with the offline clustering methods, which cannot characterize local regions with similar motions for complex 3D scenes well, leading to inaccurate scene flow estimation. To this end, we propose an iterative end-to-end super point based scene flow estimation framework, where the superpoints can be dynamically updated to guide the point-level flow prediction. Specifically, our framework consists of a flow guided superpoint generation module and a superpoint guided flow refinement module. In our superpoint generation module, we utilize the bidirectional flow information at the previous iteration to obtain the matching points of points and superpoint centers for soft point-to-superpoint association construction, in which the superpoints are generated for pairwise point clouds. With the generated superpoints, we first reconstruct the flow for each point by adaptively aggregating the superpoint-level flow, and then encode the consistency between the reconstructed flow of pairwise point clouds. Finally, we feed the consistency encoding along with the reconstructed flow into GRU to refine point-level flow. Extensive experiments on several different datasets show that our method can achieve promising performance. Code is available at https://***/supersyq/SPFlowNet.

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：