检索结果-内蒙古大学图书馆

Layer-Wise Feature Metric of Semantic-Pixel Matching for Few-Shot Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tang, Hao Lu, Junhao Huang, Guoheng Li, Ming Chen, Xuhang Zhong, Guo Tan, Zhengguang Li, Zinuo School of Computer Science and Technology Guangdong University of Technology Guangdong China School of Computer Science and Engineering Huizhou University Huizhou China Zhejiang Institute of Optoelectronics Jinhua China Zhejiang Key Laboratory of Intelligent Education Technology and Application Jinhua China School of Information Science and Technology Guangdong University of Foreign Studies Guangzhou China School of Physics Maths and Computing University of Western Australia WA Australia

In Few-Shot Learning (FSL), traditional metric-based approaches often rely on global metrics to compute similarity. However, in natural scenes, the spatial arrangement of key instances is often inconsistent across images. This spatial misalignment can result in mismatched semantic pixels, leading to inaccurate similarity measurements. To address this issue, we propose a novel method called the Layer-Wise Features Metric of Semantic-Pixel Matching (LWFM-SPM) to make finer comparisons. Our method enhances model performance through two key modules: (1) the Layer-Wise Embedding (LWE) Module, which refines the cross-correlation of image pairs to generate well-focused feature maps for each layer;(2)the Semantic-Pixel Matching (SPM) Module, which aligns critical pixels based on semantic embeddings using an assignment algorithm. We conducted extensive experiments to evaluate our method on four widely used few-shot classification benchmarks: miniImageNet, tieredImageNet, CUB-200-2011, and CIFAR-FS. The results indicate that LWFM-SPM achieves competitive performance across these benchmarks. Our code will be publicly available on https://***/Halo2Tang/Code-for-LWFM-SPM. Copyright © 2024, The Authors. All rights reserved.

关键词： Zero-shot learning

Zero Stability Well Predicts Performance of Convolutional Neural Networks

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Liangming Jin, Long Shang, Mingsheng Chongqing Key Laboratory of Big Data and Intelligent Computing Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences China Chongqing School University of Chinese Academy of Sciences China School of Information Science and Engineering Lanzhou University China

The question of what kind of convolutional neural network (CNN) structure performs well is fascinating. In this work, we move toward the answer with one more step by connecting zero stability and model performance. Specifically, we found that if a discrete solver of an ordinary differential equation is zero stable, the CNN corresponding to that solver performs well. We first give the interpretation of zero stability in the context of deep learning and then investigate the performance of existing first- and second-order CNNs under different zero-stable circumstances. Based on the preliminary observation, we provide a higher-order discretization to construct CNNs and then propose a zero-stable network (ZeroSNet). To guarantee zero stability of the ZeroSNet, we first deduce a structure that meets consistency conditions and then give a zero stable region of a training-free parameter. By analyzing the roots of a characteristic equation, we theoretically obtain the optimal coefficients of feature maps. Empirically, we present our results from three aspects: We provide extensive empirical evidence of different depth on different datasets to show that the moduli of the characteristic equation's roots are the keys for the performance of CNNs that require historical features;Our experiments show that ZeroSNet outperforms existing CNNs which is based on high-order discretization;ZeroSNets show better robustness against noises on the input. The source code is available at https://***/LongJin-lab/ZeroSNet. Copyright © 2022, The Authors. All rights reserved.

关键词： Convolutional neural networks

The NeRF Signature: Codebook-Aided Watermarking for Neural Radiance Fields

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Luo, Ziyuan Rocha, Anderson Shi, Boxin Guo, Qing Li, Haoliang Wan, Renjie Department of Computer Science Hong Kong Baptist University Hong Kong Institute of Computing University of Campinas Brazil State Key Laboratory of Multimedia Information Processing and National Engineering Research Center of Visual Technology School of Computer Science Peking University Beijing100871 China A*STAR Singapore Department of Electrical Engineering City University of Hong Kong Hong Kong

Neural Radiance Fields (NeRF) have been gaining attention as a significant form of 3D content representation. With the proliferation of NeRF-based creations, the need for copyright protection has emerged as a critical issue. Although some approaches have been proposed to embed digital watermarks into NeRF, they often neglect essential model-level considerations and incur substantial time overheads, resulting in reduced imperceptibility and robustness, along with user inconvenience. In this paper, we extend the previous criteria for image watermarking to the model level and propose NeRF Signature, a novel watermarking method for NeRF. We employ a Codebook-aided Signature Embedding (CSE) that does not alter the model structure, thereby maintaining imperceptibility and enhancing robustness at the model level. Furthermore, after optimization, any desired signatures can be embedded through the CSE, and no fine-tuning is required when NeRF owners want to use new binary signatures. Then, we introduce a joint pose-patch encryption watermarking strategy to hide signatures into patches rendered from a specific viewpoint for higher robustness. In addition, we explore a Complexity-Aware key Selection (CAKS) scheme to embed signatures in high visual complexity patches to enhance imperceptibility. The experimental results demonstrate that our method outperforms other baseline methods in terms of imperceptibility and robustness. The source code is available at: https://***/luo-ziyuan/NeRF_Signature. Copyright © 2025, The Authors. All rights reserved.

关键词： Image watermarking

RIPPB: A Robust and Improved PBFT Protocol for Blockchain

学校读者我要写书评

暂无评论

RIPPB: A Robust and Improved PBFT Protocol for Blockchain

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Meng Ding Heng He Rui Qiao Xiaohu Zhou School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan China School of Computing Engineering and Built Environment Birmingham City University Birmingham UK

ISBN: (纸本)9781665409858

In this paper, we propose RIPPB, a robust and improved PBFT protocol for Blockchain, to improve the quality of service for blockchain applications. At RIPPB, we introduce reputation for nodes and use reputation as the voting weight for node consensus to go against Sybil attacks. And we design a novel improved logistic regression model to assign reputations to nodes based on their behavior involved in block generation. With this model, nodes performing Stake-grinding attacks and Bribery attacks, face high reputation penalties, which is unacceptable for malicious nodes to reduce the probability of launching these attacks. In addition, RIPPB adds a pre-commit phase to the PBFT algorithm to reduce the number of inter-node communications, thus improving the performance of the protocol. Cost-of-attack analysis shows the high robustness of our protocol. Experimental results show a 28% improvement in the throughput of our protocol over PBFT.

关键词： Industrial electronics Protocols Quality of service Throughput Robustness Blockchains Behavioral sciences

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

学校读者我要写书评

暂无评论

International Journal of Computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Computer Science Wenzhou-Kean University Wenzhou325060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore639798 Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215400 China Jarvis Research Center Tencent YouTu Lab Shenzhen518703 China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantic Segmentation

Long-Term Water Quality In Aquaculture Based on Dilated Convolution Feature Fusion and Interactive Learning

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Li, Dashe Zhao, Weijie School of Computer Science and Technology Shandong Technology and Business University Shandong Yantai264005 China Key Laboratory of Intelligent Information Processing Shandong Technology and Business University Shandong Yantai264005 China Co-innovation Center of Shandong Colleges Universities: Future Intelligent Computing Shandong Technology and Business University Shandong Yantai264005 China

The prediction of water quality parameters is of great significance for maintaining water stability and controlling marine ***, due to the temporal nature of water quality parameters and the influence of surrounding changeable environments,water quality is characterized by complex nonlinearity and *** current prediction research, there are many training parameters and poor generalization ability, and its prediction accuracy cannot be ***,this paper proposes a network model based on dilated convolution feature fusion and interactive learning(FFINet).FFINet includes parallel SCFINet and TCINet *** FFINet model can extract the trend, periodicity and irregularity information in subsequences and performs feature fusion according to local correlations in the subsequences to capture local sequence *** global sequence features are captured by the dilated convolutional network to improve long-term dependencies between the subsequences, compensating for the information loss during the separation and feature fusion processes to some extent.A dual-residual structure is added to the model architecture to prevent vanishing and exploding gradients in the prediction *** two modules are adopted in parallel to preserve irregular information in the sequence and reduce time *** oxygen concentration experiments are conducted at multiple marine ranches,and the results of the proposed model are compared with those of other deep learning models,showing that FFINet has better prediction performance than existing *** paper provides important theoretical guidance for sustainable aquaculture development. © 2022, The Authors. All rights reserved.

关键词： Convolution

Bayesian matrix factorization for visual tracking 7

学校读者我要写书评

暂无评论

Bayesian matrix factorization for visual tracking

7th IEEE International Conference on Progress in Informatics and computing, PIC 2020

作者： Wang, Jun Wang, Zhanghua Xi, Shuaibin Wang, Yuanyun Xu, Chenguang Zhang, Jun Nanchang Institute of Technology School of Information Engineering Nanchang330029 China Jiangxi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology Nanchang330099 China

ISBN: (纸本)9781728170862

Visual tracking is an open issue in computer vision. Designing an robust tracking algorithm is a challenging problem due to complicated appearance variations, such as partial occlusion, illumination variations, fast motion and out-of-plane rotation. Some tracking algorithms use a set of templates to represent a target candidate. Such representations are not robust to drastic appearance variations. In order to alleviate the influence of appearance variations and outliers, we propose a novel tracking algorithm based on a robust Bayesian matrix factorization, where sample data is decomposed into a basis matrix and a coefficient matrix. The sample data is consist of previous tracking results. A column vector is stacked by a tracking result in a frame. A target candidate is represented by a linear combinations of the columns in the basis matrix. The matrix factorization can capture the correlation between different features in sample data. Also the appearance variations in tracking processing can be learnt in matrix factorization. Experimental results demonstrate the appearance models is effective and the proposed tracker in a particle filter tracking framework is robust to outliers and complicated appearance variations. © 2020 IEEE.

关键词： Matrix algebra

Interactions between ions and water-DMSO mixed solvent

学校读者我要写书评

暂无评论

Journal of Molecular Liquids 2024年 414卷

作者： Yan, Yuyue Zhang, Jiaqi Chen, Ligang Ren, Guanhua Zhou, Lu Liu, Liyuan Zhang, Xueqian Ouyang, Chunmei Wang, Huabin Han, Jiaguang Centre for Terahertz Waves and College of Precision Instrument and Optoelectronics Engineering Tianjin University Tianjin300072 China Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing400714 China Hebei Key Laboratory of Physics and Energy Technology Department of Mathematics and Physics North China Electric Power University Baoding071003 China Medical Science and Technology Innovation Center Shandong First Medical University Shandong Academy of Medical Sciences Shandong Jinan250021 China Guangxi Key Laboratory of Optoelectronic Information Processing School of Optoelectronic Engineering Guilin University of Electronic Technology Guilin541004 China

As a classical cryoprotectant, the cryoprotective mechanism of DMSO remains elusive. In this study, we explore the capability of DMSO to hinder the formation of salt hydrates in the water-DMSO-salt ternary system, given that, at low temperatures, salt hydrates are more prone to attach to surfaces than ice, potentially preventing mechanical damage to cell membranes. Using THz-TDS, we validate the ability of DMSO to suppress salt hydrate formation. Combining IR spectroscopy, we conclude that this inhibitory effect partly arises from the interaction between cations and the S=O groups in DMSO molecules. Semi-quantitative DR spectroscopy provides similar insights from the perspective of molecular reorientation dynamics. Taking into account the intensified ion-dipole interactions due to a higher ion surface charge density, we focus our discussion on MgCl2, while NaCl, more akin to biological processes, serves as a reference. Delving into the specific molecular mechanisms, we put forth conjectures regarding the direct interaction between cations and DMSO or the indirect influence of cations on DMSO through the modulation of hydrogen bonds in surrounding water molecules. These research findings contribute to a better understanding and mastery of cryoprotectants. © 2024 Elsevier B.V.

关键词： Terahertz spectroscopy

OpenAPMax: Abnormal Patterns-based Model for Real-World Alzheimer’s Disease Diagnosis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Huang, Yunyou Guan, Xianglong Lu, Xiangjiang Liang, Xiaoshuang Miao, Xiuxia Xie, Jiyue Liu, Wenjing Ma, Li Tang, Suqin Zhang, Zhifei Zhan, Jianfeng The Key Lab of Education Blockchain and Intelligent Technology Ministry of Education The Guangxi Key Laboratory of Multi-Source Information Mining and Security Guangxi Normal University China The School of Electronic and Information Engineering School of Integrated Circuits Guangxi Normal University Guilin530015 China The Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University China The State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100086 China The Department of Physiology and Pathophysiology Capital Medical University Beijing100069 China The Guilin Medical University Guilin541001 China The Guangxi Key Lab of Multi-Source Information Mining & Security China

Alzheimer’s disease (AD) cannot be reversed, but early diagnosis will significantly benefit patients’ medical treatment and care. In recent works, AD diagnosis has the primary assumption that all categories are known a prior—a closed-set classification problem, which contrasts with the open-set recognition problem. This assumption hinders the application of the model in natural clinical settings. Although many open-set recognition technologies have been proposed in other fields, they are challenging to use for AD diagnosis directly since 1) AD is a degenerative disease of the nervous system with similar symptoms at each stage, and it is difficult to distinguish from its pre-state, and 2) diversified strategies for AD diagnosis are challenging to model uniformly. In this work, inspired by the concerns of clinicians during diagnosis, we propose an open-set recognition model, OpenAPMax, based on the anomaly pattern to address AD diagnosis in real-world settings. OpenAPMax first obtains the abnormal pattern of each patient relative to each known category through statistics or a literature search, clusters the patients’ abnormal pattern, and finally, uses extreme value theory (EVT) to model the distance between each patient’s abnormal pattern and the center of their category and modify the classification probability. We evaluate the performance of the proposed method with recent open-set recognition, where we obtain state-of-the-art results. Copyright © 2023, The Authors. All rights reserved.

关键词： Patient treatment