检索结果-内蒙古大学图书馆

17th Asian Conference on Computer Vision, ACCV 2024

作者： Jiao, Siyi Zeng, Wenzheng Gao, Changxin Sang, Nong Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9789819609710

Interactive portrait matting refers to extracting the soft portrait from a given image that best meets the user’s intent through their inputs. Existing methods often underperform in complex scenarios, mainly due to three factors. (1) Most works apply a tightly coupled network that directly predicts matting results, lacking interpretability and resulting in inadequate modeling. (2) Existing works are limited to a single type of user input, which is ineffective for intention understanding and also inefficient for user operation. (3) The multi-round characteristics have been under-explored, which is crucial for user interaction. To alleviate these limitations, we propose DFIMat, a decoupled framework that enables flexible interactive matting. Specifically, we first decouple the task into 2 sub-ones: localizing target instances by understanding scene semantics and the flexible user inputs, and conducting refinement for instance-level matting. We observe a clear performance gain from decoupling, as it makes sub-tasks easier to learn, and the flexible multi-type input further enhances both effectiveness and efficiency. DFIMat also considers the multi-round interaction property, where a contrastive reasoning module is designed to enhance cross-round refinement. Another limitation for multi-person matting task is the lack of training data. We address this by introducing a new synthetic data generation pipeline that can generate much more realistic samples than previous arts. A new large-scale dataset SMPMat is subsequently established. Experiments verify the significant superiority of DFIMat. With it, we also investigate the roles of different input types, providing valuable principles for users. Our code and dataset will be public. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Integrating the dynamic optimization strategies of Adam and AdaGrad 13

Integrating the dynamic optimization strategies of Adam and ...

引用

13th International Conference on intelligent control and Information processing, ICICIP 2025

作者： Huang, Zhixuan Luo, Xiaonan Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin China

ISBN: (纸本)9798331516147

Artificial intelligence (AI) has been a key research area since the 1950s, initially focused on using logic and reasoning to create systems that understand language, control robots, and offer expert advice. With the rise of big data and deep learning, AI has advanced in applications like recommendation systems, image recognition, and machine translation, primarily through optimizing loss functions in deep neural networks to improve accuracy and reduce training *** descent is the core optimization method but faces challenges like slow convergence and local minima. To overcome these, algorithms like Momentum, AdaGrad, RMSProp, Adadelta, Adam, and Nadam have been developed, introducing momentum and adaptive learning rates to accelerate convergence. This paper presents a new optimization algorithm that combines the strengths of Adam and AdaGrad, offering better adaptability to different learning rates. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Attitude Detection Model Based on Improved High Resolution Network 13

Attitude Detection Model Based on Improved High Resolution N...

引用

13th International Conference on intelligent control and Information processing, ICICIP 2025

作者： Shi, Dongliang Li, Yuhan Luo, Xiaonan Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin China

ISBN: (纸本)9798331516147

2D human posture detection, a challenging task in computer vision, aims to identify human body key points and is widely used. The HRNet model performs well in human key point detection. To enhance its effect in human posture detection and address suboptimal prediction accuracy in interactive occlusion, this paper combines HRNet and Transformer. The Transformer's attention mechanism handles complex image dependencies, and HRNet obtains image features. By embedding the Transformer's attention mechanism module into HRNet's backbone network and integrating a self-designed double-branch structure and convolutional attention module, an improved HRNet model is constructed. The DConvAttention module processes extracted features, and the convolutional attention mechanism balances image frequencies, enhancing processing and discrimination of complex backgrounds and occlusions and improving detection performance. In tests of complex and occluded images, the improved HRNet Local model has a 4.2M parameter increase, 0.5G FLOPs increase, and 1% mAP increase compared to HRNet, with the same recall rate. The improved HRNet Global model has 11M more parameters, 1.1G more FLOPs, a similar or slightly higher recall rate, and a 2% higher mAP value. © 2025 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Distributed Fixed-Time Dynamic Event-Triggered Algorithm for Economic Dispatch in Smart Grids Over Directed Networks

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2025年 3卷 173-180页

作者： Ji, Lianghao Liu, Qi Zhang, Cuijuan Yang, Shasha Li, Huaqing Chongqing University of Posts and Telecommunications Chongqing Key Laboratory of Image Cognition Chongqing400065 China Southwest University Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Chongqing400715 China

A fresh algorithm is presented for tackling the economic dispatch problem (EDP) in smart grids with directed network topology. This algorithm is based on distributed consensus and aims to minimize the total cost of power generation while maintaining the balance between power supply and demand and complying with generation capacity restrictions. Even with unknown and bounded disturbances, this algorithm can achieve optimal dispatch within a fixed time. In addition, an expected settling time can be calculated, which is a more accurate upper bound estimation that relies solely on the generators' initial conditions. Meanwhile, the dynamic event-triggered strategy is employed for reducing sampling frequency and network resource consumption. Finally, numerical simulations validate the effectiveness of the designed algorithm. © 2024 IEEE.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

A High-Efficiency Doppler Frequency Parallel Acquisition Method Based on A-GPS

A High-Efficiency Doppler Frequency Parallel Acquisition Met...

引用

2025 International Technical Meeting of The Institute of Navigation, ITM 2025

作者： Jia, Qiongqiong Zhao, Yuchen China Civil Aviation University Tianjin Key Laboratory of Intelligent Signal and Image Processing Tianjin300000 China

ISBN: (纸本)9780936406404

Acquisition is a key step for Global Navigation Satellite System (GNSS) receivers to achieve positioning and other functions, and the acquisition speed directly affects the Time To First Fix (TTFF) of the receiver. The acquisition process typically involves generating local reference signals for all satellite signals and matching them with the received signals. For each satellite signal, this matching involves searching in the unknown Doppler frequency and code phase space. Traditional parallel acquisition methods can only perform parallel processing in either the Doppler frequency dimension or the code phase dimension, and still require linear search in the other dimension. To reduce the computational load and improve the acquisition efficiency, this paper proposes a high-efficiency parallel acquisition method based on A-GPS. This method utilizes the auxiliary information provided by the A-GPS base station to convert the linear search of visible star signals in Doppler frequency into a single parallel search in the Doppler search space, thereby reducing the computational load required for acquisition to 1/N of the conventional method, where N is the total number of frequency bands searched. The paper verifies through theoretical analysis and experimental results that the proposed method can significantly improve acquisition efficiency while ensuring acquisition performance. © 2025 Proceedings of the International Technical Meeting of The Institute of Navigation, ITM. All rights reserved.

关键词： Global positioning system

来源：评论

学校读者我要写书评

暂无评论

Improved LYT-Net image Enhancement Model for Underwater Robotics Vision 13

Improved LYT-Net Image Enhancement Model for Underwater Robo...

引用

13th International Conference on intelligent control and Information processing, ICICIP 2025

作者： Li, Wenjie Luo, Xiaonan Guangxi Key Laboratory of Image and Graphic Intelligent Processing Güttin University of Electronic Technology Guilin China

ISBN: (纸本)9798331516147

—Although underwater robots can replace humans to explore the ocean which is rich in resources but fraught with unknown risks, there are phenomena such as monotonous colors, complex backgrounds and uneven illumination in the underwater environment, which makes it difficult for underwater robots relying on the optical vision system to obtain clear images and is not conducive to subsequent recognition and grasping tasks. In this paper, relevant preparatory work was carried out for underwater image processing applicable to underwater robots, and the underwater imaging model as well as other underwater image processing models were analyzed. Subsequently, a method combining the underwater physical model and the dark channel image was proposed to improve LYT-Net, enabling it to be applicable to mobile embedded devices such as underwater robots. After training and testing on the real datasets LSUI and UIEB, our model achieved good results on important indicators such as PSNR and SSIM. Meanwhile, it met the theoretical requirements of mobile devices in terms of FLOPs, Params and single-frame processing time, which verified its robustness and real-time processing ability in various underwater environments. © 2025 IEEE.

关键词： Underwater imaging

来源：评论

学校读者我要写书评

暂无评论

A Novel Optimal Distributed Nonlinear Filter for Simultaneous State and Unknown Input Estimation in Multi-Sensor Networks

引用

IEEE Signal processing Letters 2025年 32卷 1745-1749页

作者： Ding, Bo Wei, Yuanchu Zhang, Enze Fang, Huajing Yangzhou University College of Information Engineering Yangzhou225127 China Huazhong University of Science and Technology School of Automation Key Laboratory of Image Processing and Intelligent Control Ministry of Education Wuhan430074 China

This letter investigates the problem of simultaneous state and unknown input estimation in a nonlinear system within multi-sensor networks. To avoid the linearization errors caused by existing methods, such as statistical linear regression techniques and first-order Taylor approximation, a novel optimal distributed nonlinear filter is proposed based on the unscented transformation and unbiased minimum variance criterion. First, the unscented transformation is applied to generate predicted estimates and the covariance matrix. Second, input and state estimation are carried out in an unbiased minimum variance manner using measurements from the nonlinear system. Third, a distributed strategy leveraging average consensus is employed to incorporate local estimates from neighboring sensors. Finally, simulation results confirm that the proposed method achieves substantial improvements in estimation accuracy compared to existing methods. © 1994-2012 IEEE.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

Underwater image Restoration Based on Improved SwinIR Model 13

Underwater Image Restoration Based on Improved SwinIR Model

引用

13th International Conference on intelligent control and Information processing, ICICIP 2025

作者： Lu, Yanke Li, Fang Luo, Xiaonan Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin541004 China School of Computer Science and Information Security Guilin University of Electronic Technology Guilin541004 China Guilin541004 China Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin541004 China

ISBN: (纸本)9798331516147

image restoration is a classic foundational visual task, aimed at recovering damaged images, such as those affected by compression, blurring, or noise, to high-definition clarity. Although current image enhancement techniques largely rely on convolutional neural networks, the efficient capabilities of Transformers demonstrated in complex visual tasks have not yet been extensively applied in the field of image restoration. This paper presents an optimized image restoration strategy based on the SwinIR model, which is applied to the task of underwater image restoration. In the shallow feature extraction phase, deep separable convolution technology is employed;while in the deep feature extraction phase, a Spatial Gated Feedforward Network (SGFN) is utilized. The experimental data suggests that the novel approach surpasses conventional techniques in effectiveness and has effectively minimized both the model's parameter quantity and computational demands. © 2025 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Complementary Learning Subnetworks towards Parameter-Efficient Class-Incremental Learning

引用

IEEE Transactions on Knowledge and Data Engineering 2025年第6期37卷 3240-3252页

作者： Li, Depeng Zeng, Zhigang Dai, Wei Suganthan, Ponnuthurai Nagaratnam Huazhong University of Science and Technology School of Artificial Intelligence and Automation China Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Wuhan430074 China China University of Mining and Technology School of Information and Control Engineering Xuzhou221116 China Qatar University KINDI Center for Computing Research College of Engineering Doha Qatar

In the scenario of class-incremental learning (CIL), deep neural networks have to adapt their model parameters to non-stationary data distributions, e.g., the emergence of new classes over time. To mitigate the catastrophic forgetting phenomenon, typical CIL methods either cumulatively store exemplars of old classes for retraining model parameters from scratch or progressively expand model size as new classes arrive, which, however, compromises their practical value due to little attention paid to parameter efficiency. In this paper, we contribute a novel solution, effective control of the parameters of a well-trained model, by the synergy between two complementary learning subnetworks. Specifically, we integrate one plastic feature extractor and one analytical feed-forward classifier into a unified framework amenable to streaming data. In each CIL session, it achieves non-overwritten parameter updates in a cost-effective manner, neither revisiting old task data nor extending previously learned networks;Instead, it accommodates new tasks by attaching a tiny set of declarative parameters to its backbone, in which only one matrix per task or one vector per class is kept for knowledge retention. Experimental results on a variety of task sequences demonstrate that our method achieves competitive results against state-of-the-art CIL approaches, especially in accuracy gain, knowledge transfer, training efficiency, and task-order robustness. Furthermore, a graceful forgetting implementation on previously learned trivial tasks is empirically investigated to make its non-growing backbone (i.e., a model with limited network capacity) suffice to train on more incoming tasks. © 2025 IEEE. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Curve-Suppression-Based Event-Triggered Mechanisms for Quasi-Synchronization of Fuzzy Delayed Neural Networks on Time Scales

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3174-3187页

作者： Wan, Peng Zhou, Yufeng Zeng, Zhigang Lai, Jingang Wuhan University of Science and Technology School of Artificial Intelligence and Automation Engineering Research Center of Metallurgical Automation and Measurement Technology Wuhan430081 China Huazhong University of Science and Technology School of Artificial Intelligence and Automation Key Laboratory of Image Processing and Intelligent control of Education Ministry of China Wuhan430074 China

The vast majority of published event-triggered mechanisms (ETMs) are constructed based on measurement errors, which introduces a problem naturally that they are updated when the measurement errors exceed the thresholds although the current obtained sampling states can make systems converge well. With this problem in mind, we redesign ETMs for quasi-synchronization of T-S fuzzy neural networks (FNNs) with time delays on time scales. First, a novel ETM is designed for continuous-time FNNs with time-varying delays to achieve quasi-synchronization, with which synchronization errors is suppressed to globally exponentially converge to a ball. Second, we introduce the ETM for continuous-time FNNs to discrete-time FNNs, owing to the existence of discrete-time states, the Lypunov function of synchronization errors run over the exponentially decay curve, but it can be suppressed to evolve under another exponentially decay curve. Third, for FNNs on time scales with constant and time-varying delays, we estimate the forward jump operator of the Lyapunov functions and design ETMs to guarantee that the Lypunov functions evolve under the exponentially decay curves, so quasi-synchronization can be achieved. Last but not least, we prove that Zeno behavior will not happen and four numerical examples are introduced to verify the validity and the superiority of the proposed ETMs in reducing information transmission. © 2013 IEEE.

关键词： Continuous time systems

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：