检索结果-内蒙古大学图书馆

Reinforcement learning of non-additive joint steganographic embedding costs with attention mechanism

Science China(Information Sciences) 2023年第3期66卷 273-286页

作者： Weixuan TANG Bin LI Weixiang LI Yuangen WANG Jiwu HUANG Institute of Artificial Intelligence and Blockchain Guangzhou University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen Key Laboratory of Media Security Shenzhen University Shenzhen Institute of Artificial Intelligence and Robotics for Society School of Computer Science and Cyber Engineering Guangzhou University

Image steganography is the art and science of secure communication by concealing information within digital images. In recent years, the techniques of steganographic cost learning have developed rapidly. Although the existing methods can learn satisfactory additive costs, the interplay of different pixels' embedding impacts has not been considered, so the potential of learning may not be fully exploited. To overcome this limitation, in this paper, a reinforcement learning paradigm called Jo Po L(joint policy learning) is proposed to extend the idea of additive cost learning to a non-additive situation. Jo Po L aims to capture the interactions within pixel blocks by defining embedding policies and evaluating contributions of embedding impacts on a block level rather than a pixel level. Then, a policy network is utilized to learn optimal joint embedding policies for pixel blocks through interactions with the environment. Afterwards,these policies can be converted into joint embedding costs for practical message embedding. The structure of the policy network is designed with an effective attention mechanism and incorporated with the domain knowledge derived from traditional non-additive steganographic methods. The environment is responsible for assigning rewards according to the impacts of the sampled joint embedding actions, which are evaluated by the gradient information of a neural network-based steganalyzer. Experimental results show that the proposed non-additive method Jo Po L significantly outperforms the existing additive methods against both feature-based and CNN-based steganalzyers over different payloads.

关键词： information hiding non-additive steganography steganalysis cost learning image processing

来源：评论

学校读者我要写书评

暂无评论

ESE-GAN: Zero-Shot Food Image Classification Based on Low Dimensional Embedding of Visual Features

引用

IEEE Transactions on Multimedia 2024年 27卷 1-11页

作者： Li, Gaojie Li, Yaochen Liu, Jingle Guo, Wei Tang, Wenneng Liu, Yuehu School of Software Engineering Xi'an Jiaotong University Xi'an China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China

Existing zero-shot learning based image classification methods transform the zero-shot learning problem into supervised learning by applying generative adversarial network (GAN) to synthesize visual features of unseen classes. However, the visual features generated by the generator tend to be biased towards seen classes, and the discriminator is too weak to generate high-quality image features. To solve these problems, we propose a novel zero-shot food image classification method based on low dimensional embedding of visual features. Our method applies reinforced semantic guidance to increase the discriminative ability of the model by enhancing the strong distribution of input features. Moreover, the visual space is utilized as the embedding space to reduce the bias towards seen classes by reducing the distance between semantic information and visual features in the embedding space. Finally, the feature distribution of unseen classes is further specified by improving the prototype similarity function. Extensive experiments on three food datasets and four general benchmark datasets demonstrate the effectiveness of the proposed method. IEEE

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Grasp Detection with Hierarchical Multi-Scale Feature Fusion and Inverted Shuffle Residual

引用

Tsinghua Science and Technology 2024年第1期29卷 244-256页

作者： Wenjie Geng Zhiqiang Cao Peiyu Guan Fengshui Jing Min Tan Junzhi Yu State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China Department of Advanced Manufacturing and Robotics College of EngineeringPeking UniversityBeijing 100871China

Grasp detection plays a critical role for robot *** pixel-wise grasp detection networks with encoder-decoder structure receive much attention due to good accuracy and ***,they usually transmit the high-level feature in the encoder to the decoder,and low-level features are *** is noted that low-level features contain abundant detail information,and how to fully exploit low-level features remains ***,the channel information in high-level feature is also not well ***,the performance of grasp detection is *** solve these problems,we propose a grasp detection network with hierarchical multi-scale feature fusion and inverted shuffle *** low-level and high-level features in the encoder are firstly fused by the designed skip connections with attention module,and the fused information is then propagated to corresponding layers of the decoder for in-depth feature *** a hierarchical fusion guarantees the quality of grasp ***,an inverted shuffle residual module is created,where the high-level feature from encoder is split in channel and the resultant split features are processed in their respective *** such differentiation processing,more high-dimensional channel information is kept,which enhances the representation ability of the ***,an information enhancement module is added before the encoder to reinforce input *** proposed method attains 98.9%and 97.8%in image-wise and object-wise accuracy on the Cornell grasping dataset,respectively,and the experimental results verify the effectiveness of the method.

关键词： grasp detection hierarchical multi-scale feature fusion skip connections with attention inverted shuffle residual

来源：评论

学校读者我要写书评

暂无评论

Out-Bore MRI-Guided Breast Biopsy Employing Machine Vision and Augmented Reality Techniques: Preliminary Study 46

Out-Bore MRI-Guided Breast Biopsy Employing Machine Vision a...

引用

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024

作者： Jeon, Sangseo Kim, Yeonhyeong Lim, Sunghwan Korea Institute of Science and Technology Center for Healthcare Robotics Artificial Intelligence and Robotics Institute Seoul02792 Korea Republic of

ISBN: (纸本)9798350371499

Magnetic resonance imaging (MRI) of the breast demonstrates superior sensitivity and specificity in distinguishing cancerous tissue. Despite the availability of in-bore MRI-guided breast biopsy, its application is restricted to a limited subset of individuals. We present a novel methodology developed for out-bore MRI-guided breast biopsy. Our approach integrates three primary functionalities: real-time pose tracking of the biopsy needle and a head-mounted display (HMD), real-time shape sensing of the breast with deformable registration, and augmented reality (AR)-assisted visualization of the MRI target and navigational information. Comprehensive experiments were performed to investigate the core performance metrics of the method, including targeting accuracy, update rate of the deformable registration, and data refresh rate and latency of the AR visualization. The mean targeting accuracy was 4.55 ± 1.11mm. The average update rate for deformable registration was 42.76 ± 13.96ms. Furthermore, the data refresh rate and latency associated with AR visualization were recorded as 59 ± 23f ps and 192 ± 73ms, respectively. The experimental findings underscore the system's promising potential in facilitating successful outcomes for out-bore MRI-guided breast *** relevance - The proposed system facilitates out-bore MRI-guided breast biopsy, enhancing accessibility to MRI-guided breast biopsy for patients. This expansion is anticipated to significantly elevate the detection rate of breast cancer © 2024 IEEE.

关键词： Biopsy

来源：评论

学校读者我要写书评

暂无评论

Traditional Transformation Theory Guided Model for Learned Image Compression

Traditional Transformation Theory Guided Model for Learned I...

引用

2024 IEEE International Conference on Consumer Electronics, ICCE 2024

作者： Li, Zhiyuan Ge, Chenyang Li, Shun Xi'An Jiaotong University Institute of Artificial Intelligence and Robotics Xi'an710049 China

ISBN: (纸本)9798350324136

Recently, many deep image compression methods have been proposed and achieved remarkable performance. However, these methods are dedicated to optimizing the compression performance and speed at medium and high bitrates, while research on ultra low bitrates is limited. In this work, we propose a ultra low bitrates enhanced invertible encoding network guided by traditional transformation theory, experiments show that our codec outperforms existing methods in both compression and reconstruction performance. Specifically, we introduce the Block Discrete Cosine Transformation to model the sparsity of features and employ traditional Haar transformation to improve the reconstruction performance of the model without increasing the bitstream cost. © 2024 IEEE.

关键词： Deep image compression invertible codec traditional transformation theory ultra low bitrates

来源：评论

学校读者我要写书评

暂无评论

EATNet: Efficient Axial Transformer Network for End-to-end Autonomous Driving 27

EATNet: Efficient Axial Transformer Network for End-to-end A...

引用

27th IEEE International Conference on Intelligent Transportation Systems, ITSC 2024

作者： Chen, Weihuang Kong, Fanjie Chen, Liming Wang, Shen'ao Wang, Zhiping Sun, Hongbin Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9798331505929

In recent years, end-to-end autonomous driving has garnered significant attention from researchers and has witnessed rapid advancements. However, existing methods en-counter challenges such as high computational demands, slow training and inference speeds, which hinder their real-world deployment. To tackle this issue, we introduce the Efficient Axial Transformer Network (EATNet), a lightweight multi-modal autonomous driving framework based on cross-axial Transformers. By effectively integrating lidar and multi-view RGB features, this model utilizes an enhanced lightweight cross-axial Transformer to minimize model size and computational requirements. Extensive experiments demonstrate that EATNet, with only a quarter of the parameters of comparable multi-modal models, achieves competitive or even superior performance on the closed-loop CARLA simulator compared to other baselines. © 2024 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Modeling and precise tracking control of spatial bending pneumatic softactuators

引用

Biomimetic intelligence & robotics 2024年第4期4卷 92-98页

作者： Yize Ma Qingxiang Wu Zehao Qiu Yongchun Fang Ning Sun Institute of Robotics and Automatic Information Systems(IRAIS) College of Artificial IntelligenceNankai UniversityTianjin 300350China Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute of Nankai UniversityShenzhen 518083China

In recent years,a variety of pneumatic soft actuators(PSAs)have been proposed due to the develop-ment of soft robots in biomimetic robots,medical devices,*** the same time,the modeling and control of PSAs remains an open *** this paper,a spatial bending pneumatic soft actuator(SBPSA)modeling method based on the Prandtl-Ishlinskii(PI)model is proposed,and the inverse model is designed to compensate for hysteresis ***,an adaptive feedback controller combined with a hysteresis compensator is proposed for the precise control and tracking of ***,an experimental platform is built,and experimental results demonstrate the effectiveness of the proposed method for precise tracking.

关键词： Spatial bending Hysteresis compensation Prandtl-Ishlinskii modeling Adaptive feedback control

来源：评论

学校读者我要写书评

暂无评论

Human-like dexterous manipulation for anthropomorphic five-fingered hands:A review

引用

Biomimetic intelligence & robotics 2025年第1期5卷 35-57页

作者： Yayu Huang Dongxuan Fan Haonan Duan Dashun Yan Wen Qi Jia Sun Qian Liu Peng Wang Institute of Automation Chinese Academy of SciencesBeijing 100190China School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science and InnovationChinese Academy of SciencesHong Kong 999077China

Humans excel at dexterous manipulation;however,achieving human-level dexterity remains a significant challenge for *** breakthroughs in the design of anthropomorphic robotic hands,as well as advancements in visual and tactile perception,have demonstrated significant advantages in addressing this ***,coping with the inevitable uncertainty caused by unstructured and dynamic environments in human-like dexterous manipulation tasks,especially for anthropomorphic five-fingered hands,remains an open *** this paper,we present a focused review of human-like dexterous manipulation for anthropomorphic five-fingered *** begin by defining human-like dexterity and outlining the tasks associated with human-like robot dexterous ***,we delve into anthropomorphism and anthropomorphic five-fingered hands,covering definitions,robotic design,and evaluation ***,we review the learning methods for achieving human-like dexterity in anthropomorphic five-fingered hands,including imitation learning,reinforcement learning and their ***,we discuss the existing challenges and propose future research *** review aims to stimulate interest in scientific research and future applications.

关键词： Anthropomorphic five-fingered hands Human-like dexterous manipulation Robot learning Review

来源：评论

学校读者我要写书评

暂无评论

Learning Residual Model of Model Predictive Control via Random Forests for Autonomous Driving 27

Learning Residual Model of Model Predictive Control via Rand...

引用

27th IEEE International Conference on Intelligent Transportation Systems, ITSC 2024

作者： Zhao, Kang Xue, Jianru Meng, Xiangning Li, Gengxin Wu, Mengsen Institute of Artificial Intelligence and Robotics College of Artificial Intelligence Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Xi'an China

ISBN: (纸本)9798331505929

One major issue in learning-based model predictive control (MPC) for autonomous driving is the contradiction between the system model's prediction accuracy and computation efficiency. The more situations a system model covers, the more complex it is, along with highly nonlinear and nonconvex properties. These issues make the optimization too complicated to solve and render real-time control impractical. To address these issues, we propose a hierarchical learning residual model which leverages random forests and linear regression. The learned model consists of two levels. The low level uses linear regression to fit the residues, and the high level uses random forests to switch different linear models. Meanwhile, we adopt the linear dynamic bicycle model with error states as the nominal model. The switched linear regression model is added to the nominal model to form the system model. It reformulates the learning-based MPC as a quadratic program (QP) problem, and optimization solvers can effectively solve it. Experimental path tracking results show that the driving vehicle's prediction accuracy and tracking accuracy are significantly improved compared with the nominal MPC. Compared with the state-of-the-art Gaussian process-based nonlinear model predictive control (GP-NMPC), our method gets better performance on tracking accuracy while maintaining a lower computation consumption. © 2024 IEEE.

关键词： Predictive control systems

来源：评论

学校读者我要写书评

暂无评论

Efficient Nonnegative Tensor Decomposition Using Alternating Direction Proximal Method of Multipliers

引用

Chinese Journal of Electronics 2024年第5期33卷 1308-1316页

作者： Deqing WANG Guoqiang HU State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Key Laboratory of Marine Robotics Liaoning Province College of Artificial Intelligence Dalian Maritime University

Nonnegative CANDECOMP/PARAFAC（NCP） tensor decomposition is a powerful tool for multiway signal processing. The alternating direction method of multipliers（ADMM） optimization algorithm has become increasingly popular for solving tensor decomposition problems in the block coordinate descent framework. However,the ADMM-based NCP algorithm suffers from rank deficiency and slow convergence for some large-scale and highly sparse tensor data. The proximal algorithm is preferred to enhance optimization algorithms and improve convergence properties. In this study, we propose a novel NCP algorithm using the alternating direction proximal method of multipliers（ADPMM） that consists of the proximal algorithm. The proposed NCP algorithm can guarantee convergence and overcome the rank deficiency. Moreover, we implement the proposed NCP using an inexact scheme that alternatively optimizes the subproblems. Each subproblem is optimized by a finite number of inner iterations yielding fast computation speed. Our NCP algorithm is a hybrid of alternating optimization and ADPMM and is named *** experimental results on synthetic and real-world tensors demonstrate the effectiveness and efficiency of our proposed algorithm.

关键词： Tensors Neuroscience Signal processing algorithms Linearization techniques Signal processing Fluorescence Data processing

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：