检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Lianyu Pang Jian Yin Haoran Xie Qiping Wang Qing Li Xudong Mao Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Lingnan University East China Normal University The Hong Kong Polytechnic University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Recently, there has been a surge in face personalization techniques, benefiting from the advanced capabilities of pretrained text-to-image diffusion models. Among these, a notable method is Textual Inversion, which generates personalized images by inverting given images into textual embeddings. However, methods based on Textual Inversion still struggle with balancing the tradeoff between reconstruction quality and editability. In this study, we examine this issue through the lens of initialization. Upon closely examining traditional initialization methods, we identified a significant disparity between the initial and learned embeddings in terms of both scale and orientation. The scale of the learned embedding can be up to 100 times greater than that of the initial embedding. Such a significant change in the embedding could increase the risk of overfitting, thereby compromising the editability. Driven by this observation, we introduce a novel initialization method, termed Cross Initialization, that significantly narrows the gap between the initial and learned embeddings. This method not only improves both reconstruction and editability but also reduces the optimization steps from 5,000 to 320. Furthermore, we apply a regularization term to keep the learned embedding close to the initial embedding. We show that when combined with Cross Initialization, this regularization term can effectively improve editability. We provide comprehensive empirical evidence to demonstrate the superior performance of our method compared to the baseline methods. Notably, in our experiments, Cross Initialization is the only method that successfully edits an individual's facial expression. Additionally, a fast version of our method allows for capturing an input image in roughly 26 seconds, while surpassing the baseline methods in terms of both reconstruction and editability. Code is available at https://***/lyuPang/CrossInitialization.

关键词： Computer vision Face recognition Computational modeling Text to image Diffusion models Surges Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Lightweight Fine-Tuning a Pretrained Protein Language Model for Protein Secondary Structure Prediction

SSRN

引用

SSRN 2023年

作者： Yang, Wei Liu, Chun Li, Zheng Henan Key Laboratory of Big Data Analysis and Processing Henan Engineering Laboratory of Spatial Information Processing School of Computer and Information Engineering Henan University Kaifeng475004 China

Pretrained large-scale protein language models, such as ESM-1b and ProtTrans, are becoming the fundamental infrastructure for various protein-related biological modeling tasks. Existing works use mainly pretrained protein language models in feature extraction. However, the knowledge contained in the embedding features directly extracted from a pretrained model is task-agnostic. To obtain task-specific feature representations, a reasonable approach is to fine-tune a pretrained model based on labeled datasets from downstream tasks. To this end, we investigate the fine-tuning of a given pretrained protein language model for protein secondary structure prediction tasks. Specifically, we propose a novel end-to-end protein secondary structure prediction framework involving the lightweight fine-tuning of a pretrained model. The framework first introduces a few new parameters for each transformer block in the pretrained model, then updates only the newly introduced parameters, and then keeps the original pretrained parameters fixed during training. Extensive experiments on seven test sets, namely, CASP12, CASP13, CASP14, CB433, CB634, TEST2016, and TEST2018, show that the proposed framework outperforms existing predictors and achieves new state-of-the-art prediction performance. Furthermore, we also experimentally demonstrate that lightweight fine-tuning significantly outperforms full model fine-tuning and feature extraction in enabling models to predict secondary structures. Further analysis indicates that only a few top transformer blocks need to introduce new parameters, while skipping many lower transformer blocks has little impact on the prediction accuracy of secondary structures. Our code and datasets are publicly available at https://***/fengtuan/LIFT_SS. © 2023, The Authors. All rights reserved.

关键词： Proteins

来源：评论

学校读者我要写书评

暂无评论

ALEDAR: An Attentions-based Encoder-Decoder and Autoregressive model for workload Forecasting of Cloud data Center 25

ALEDAR: An Attentions-based Encoder-Decoder and Autoregressi...

引用

25th IEEE International Conference on Computer Supported Cooperative Work in Design, CSCWD 2022

作者： Sun, Wei Xu, Xiaolong Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Big Data Security and Intelligent Processing School of Computer Science Nanjing China

ISBN: (纸本)9781665405270

Effective workload forecasting can provide a reference for resource scheduling in cloud data centers. Compared with the normal single data center, the multi-data center has a more complicated architecture design and provides more diverse computing services. The traditional load forecasting models need too much manual intervention to set up the parameters, meanwhile, the emerging neural network methods are not sensitive to the prediction scale and cannot capture the long-term associations between input features. To address these challenges, we propose a hybrid model named Attentions-based LSTM Encoder-Decoder network and Autoregressive model (ALEDAR), which combines neural network and statistical learning methods, to analyze the linear and nonlinear characteristics of the load sequence over time in a multi-cloud data center environment. ALEDAR uses a dual attention-based Encoder-Decoder architecture to extract the relationships between historical workload data, the design can also avoid the deterioration of prediction effect caused by long-range data scale, the output layer is composed of a three-layer perceptron. Moreover, ALEDAR employs an autoregressive module to capture the linear trend of the load sequence and eliminate scale insensitivity of input and output data. The experimental results show that our approach is adaptive and can improve the performance of host workload prediction in both single cloud data center and multi-cloud data center environments. We evaluate the performance of the proposed algorithm on real-world data sets and observe consistent improvement of 9.7% - 34.2% over the state-of-the-art baselines. © 2022 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

An Adaptive DoH Encrypted Tunnel Detection Method Based on Contrastive Learning

引用

IEEE Internet of Things Journal 2025年

作者： Tong, Jiacheng Zhao, Yilin Jin, Chongju Chen, Wei Zhang, Yiting Wu, Lifa Nanjing University of Posts and Telecommunications School of Computer Science Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing210003 China

The percentage of encrypted network traffic has constantly increased as network security has been continuously improved. Attackers can, however, utilize encrypted DNS over HTTPS (DoH) to conceal their malicious traffic, which makes it more difficult to identify malicious tunnels. To address this issue, we first examine the encryption features of DoH tunnel traffic. Due to the incapability of current detection techniques to properly fuse traffic attributes, a fusion learning-based method is proposed to detect DoH encrypted tunnel traffic. At the same time, we discover that the DoH traffic samples may exhibit concept drift. As a result, we present a concept drift detection approach based on a contrastive sparse autoencoder. In addition to the above method, a model retraining strategy is also suggested to improve the model’s capacity to identify new DoH encrypted tunnel traffic while reducing its reliance on expert label data. This strategy involves incrementally training the model using as few samples as possible. Experiments demonstrate that the proposed method can significantly enhance detection performance. When 7% of drift samples are used during incremental training, the detection accuracy of the model recovers from 74.02% to 99.96%. © 2014 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Carbon emission reduction enabled by informatization construction: an analysis of spatial effects based on China’s experience

引用

Environmental Science and Pollution Research 2024年第24期31卷 35595-35608页

作者： He, Zhuo Ma, Shaojuan Deng, Zixin Meng, Ying College of Business Administration Nanchang Institute of Technology Jiangxi Nanchang330032 China School of Mathematics and Information Science Northern Minzu University Ningxia Yinchuan750021 China Ningxia Key Laboratory of Intelligent Information and Big Data Processing Ningxia Yinchuan750021 China School of Statistics Capital University of Economics and Business Beijing100071 China

The "dual-carbon" objective presents a huge challenge for China and the world, with profound implications for the advancement of China’s eco-friendly economy. Additionally, informatization development has a significant impact on the level of carbon emissions in both local and neighbouring regions. Therefore, we employ panel data from 30 provinces in China spanning the years 2012 to 2021, and use the Kernel density estimate and Moran’s index to explore informatization level and carbon emissions space agglomeration characteristics. We elucidate the nonlinear relationship and heterogeneity between informatization improvement and carbon emissions based on the spatial Durbin model. The primary findings are as follows. Firstly, we discover a distinct spatial clustering phenomenon which the informatization level is high in coastal areas and low in inland areas, whereas carbon emissions are low in the south and high in the north. Secondly, the effect of the informatization level on carbon emissions is shown as a U-shaped and non-linear correlation, signifying inhibitory and subsequently promoting phases. Thirdly, we reveal the negative influence on carbon emissions caused by spatial lag terms of the informatization level, and find that a higher local informatization level will have an inhibitory effect on carbon emissions in neighbouring areas. Finally, there is a spatial heterogeneity in the impact of the informatization level on carbon emissions, which presents the U-shaped relation between informatization level and carbon emissions varies across the North-South subregion and the three major economic subregion of China. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Informatization

来源：评论

学校读者我要写书评

暂无评论

TMFNet: Two-Stream Multi-Channels Fusion Networks for Color Image Operation Chain Detection

arXiv

引用

arXiv 2024年

作者： Niu, Yakun Tan, Lei Zhang, Lei Zuo, Xianyu Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng475000 China School of Computer and Information Engineering Henan University Kaifeng475000 China

Image operation chain detection techniques have gained increasing attention recently in the field of multimedia forensics. However, existing detection methods suffer from the generalization problem. Moreover, the channel correlation of color images that provides additional forensic evidence is often ignored. To solve these issues, in this article, we propose a novel two-stream multi-channels fusion networks for color image operation chain detection in which the spatial artifact stream and the noise residual stream are explored in a complementary manner. Specifically, we first propose a novel deep residual architecture without pooling in the spatial artifact stream for learning the global features representation of multi-channel correlation. Then, a set of filters is designed to aggregate the correlation information of multi-channels while capturing the low-level features in the noise residual stream. Subsequently, the high-level features are extracted by the deep residual model. Finally, features from the two streams are fed into a fusion module, to effectively learn richer discriminative representations of the operation chain. Extensive experiments show that the proposed method achieves state-of-the-art generalization ability while maintaining robustness to JPEG compression. The source code used in these experiments will be released at https://***/LeiTan98/TMFNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

UAV 3D Path Planning Based on Improved Chimp Optimization Algorithm

引用

Computers, Materials & Continua 2025年第6期83卷 5679-5698页

作者： Wenli Lei Xinghao Wu KunJia Jinping Han School of Physics and Electronic Information Yan’an UniversityYan’an716000China Shaanxi Key Laboratory of Intelligent Processing for Big Energy Data Yan’an716000China

Aiming to address the limitations of the standard Chimp Optimization Algorithm(ChOA),such as inadequate search ability and susceptibility to local optima in Unmanned Aerial Vehicle(UAV)path planning,this paper proposes a three-dimensional path planning method for UAVs based on the Improved Chimp Optimization Algorithm(IChOA).First,this paper models the terrain and obstacle environments spatially and formulates the total UAV flight cost function according to the constraints,transforming the path planning problem into an optimization problem with multiple ***,this paper enhances the diversity of the chimpanzee population by applying the Sine chaos mapping strategy and introduces a nonlinear convergence factor to improve the algorithm’s search accuracy and convergence ***,this paper proposes a dynamic adjustment strategy for the number of chimpanzee advance echelons,which effectively balances global exploration and local exploitation,significantly optimizing the algorithm’s search *** validate the effectiveness of the IChOA algorithm,this paper conducts experimental comparisons with eight different intelligent *** experimental results demonstrate that the IChOA outperforms the selected comparison algorithms in terms of practicality and robustness in UAV 3D path *** effectively solves the issues of efficiency in finding the shortest path and ensures high stability during execution.

关键词： UAV path planning chimp optimization algorithm chaotic mapping adaptive weighting

来源：评论

学校读者我要写书评

暂无评论

A New Semi-supervised Approach for Network Encrypted Traffic Clustering and Classification 25

A New Semi-supervised Approach for Network Encrypted Traffic...

引用

25th IEEE International Conference on Computer Supported Cooperative Work in Design, CSCWD 2022

作者： Lin, Kunda Xu, Xiaolong Jiang, Yu School of Computer Science Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing China

ISBN: (纸本)9781665405270

Encrypted network traffic classification is an essential task in modern communications, which is used in a wide range of applications, such as network resource allocation, QoS (Quality of Service), malicious detection, etc. With the continuous evolution of network technology, approaches used to classify network encrypted traffic have become increasingly complex, and model training relies heavily on a large amount of labeled data. However, the acquisition of correct and massive labeled data of network traffic in the real environment remains a major challenge in this field. On the opposite, unlabeled traffic is extremely easy to obtain in a network. Therefore, the effective use of unlabeled data is of great significance to the development of modern communications. In this paper, we propose the Sauce model, which can effectively use unlabeled information to obtain high-quality clustering space. It is composed of Aux (auxiliary network) and AE (auto encoder), where Aux is designed for collaborative training with AE, affecting the distribution of samples in the latent space generated by AE. Sauce uses t-SNE (t- distributed stochastic neighbor embedding) to perform secondary dimensionality reduction on the latent code, and then uses a clustering algorithm for cluster analysis. Besides, sauce applies the self-training technique, replacing the real labeled data with pseudo-labeled data to reduce the reliance on labeled data and improve the utilization of unannotated data. We conduct experiments on two real network datasets. The experiments show that Sauce can achieve clustering accuracies of 98.4% (VPN dataset) and 98.0% (TOR dataset), outperforming other unsupervised or semi-supervised learning methods. © 2022 IEEE.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Deep learning approaches in flow visualization

引用

Advances in Aerodynamics 2022年第1期4卷 327-340页

作者： Can Liu Ruike Jiang Datong Wei Changhe Yang Yanda Li Fang Wang Xiaoru Yuan Key Laboratory of Machine Perception(Ministry of Education) and School of AIPeking UniversityBeijingChina China Aerodynamics Research and Development Center MianyangChina National Engineering Laboratory for Big Data Analysis and Application Peking UniversityBeijingChina

With the development of deep learning(DL)techniques,many tasks in flow visualization that used to rely on complex analysis algorithms now can be replaced by DL *** reviewed the approaches to deep learning technology in flow visualization and discussed the technical benefits of these *** also analyzed the prospects of the development of flow visualization with the help of deep learning.

关键词： Deep learning Flow visualization data management Feature extraction Particle tracing

来源：评论

学校读者我要写书评

暂无评论

A Digital Twin-based Framework of Motion Control and State Monitoring for Pneumatic Muscle

引用

Recent Patents on Mechanical Engineering 2023年第3期16卷 203-213页

作者： Xie, Shenglong Liu, Wenyuan Duan, Huiru Chen, Dijian Wan, Yanjian Key Laboratory of Intelligent Manufacturing Quality Big Data Tracking and Analysis of Zhejiang Province China Ji-liang University Zhejiang Hangzhou310018 China Anhui Engineering Research Center for Intelligent Applications and Security of Industrial Internet Anhui University of Technology Anhui Ma'anshan243032 China

Introduction: The current digital twin systems usually have the drawback of high cost and complex technology, and it is necessary to develop a simple solution to reduce the cost and cycle for the development of digital twin systems, especially for small projects or systems with simple structures. Objective: A low-cost patent technology of digital twin system was proposed by taking the motion control and state monitoring system (MCSMS) of pneumatic muscle as an example. Methods: The MCSMS is developed based on the browser/server architecture. The software of 3ds Max and SolidWorks are used to make the virtual model, *** and JavaScript are applied to build the browser side. data of the physical world is collected and processed on the server side firstly, and then is sent to the browser side through HTTP communication protocol to realize data interchange between the browser and server. Results: In the roaming experiment and the experiment of motion control and state monitoring of pneumatic muscle, the MCSMS can work smoothly without obvious delay and has good real-time performance, which can realize the 3D visual monitoring of the pneumatic muscle very well. Conclusion: The experimental results indicate that the proposed method possesses the ability of good feasibility and effectiveness. © 2023 Bentham Science Publishers.

关键词： Muscle

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：