检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Zeng, Yu Zhang, Yang Liu, Jiachen Shen, Linlin Deng, Kaijun He, Weizhao Wang, Jinbao Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Semi-supervised Eye Semantic Segmentation Method Based on Iterative Learning 8

A Lightweight Semi-supervised Eye Semantic Segmentation Meth...

引用

2024 8th International Conference on Machine Vision and Information technology, CMVIT 2024

作者： Qi, Ruochen Zhang, Guanghui Zhu, Dongchen Shi, Wenjun Zhang, Xiaolin Bionic Vision System Laboratory State Key Laboratory of Transducer Technology Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai200050 China University of Chinese Academy of Sciences Beijing100049 China School of Information Science and Technology ShanghaiTech University Shanghai201210 China

Eye semantic segmentation has a significant role in promoting biometric recognition. Most works improve the accuracy through designing complicated feature learning modules that usually require a high computation cost and are prone to over-fit. In this work, we propose a lightweight eye semantic segmentation method based on iterative learning, which only requires a small amount of labeled eye image data under visible light conditions. To be specific, we first propose a pseudo-label filtering strategy for eye semantic segmentation, which leverages a semi-supervised incremental learning technique to solve the problem of label scarcity. Besides, an attention-guided lightweight high-representation module is designed to boost precision while ensuring efficiency. Experiments on the Ubiris dataset show that the proposed method achieves state-of-the-art performance, with a mIoU of 86.18%. © Published under licence by IOP Publishing Ltd.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

An Autonomous Obstacle Avoidance Path Planning Method Involving PSO for Dual-Arm Surgical Robot

An Autonomous Obstacle Avoidance Path Planning Method Involv...

引用

IEEE International Conference on Mechatronics, robotics and Automation (ICMRA)

作者： Qiao Chen Yiwei Liu Zhuo Chen Yangjunjian Zhou State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

Currently, motion planning and reasonable obstacle avoidance of surgical robots are essential research directions. Most surgical robots adopt a simple master-slave control strategy and cannot avoid obstacles autonomously. To this end, we propose an autonomous obstacle avoidance method for the dual-arm surgical robot. Firstly, the obstacle avoidance path planning is based on the artificial potential field method. We propose the splicing algorithm to make the manipulator jump out of the local minimum and improve the planning efficiency. And using the adaptive step to stable motion. Secondly, for collision detection, we use the Gilbert-Johnson-Keerthi (GJK) algorithm to calculate the distance between the obstacle and the equivalent convex body of the manipulator. A warning appears when there is a collision. Because the artificial potential field method lacks time planning, we use the time-optimized 3-5-3 polynomial interpolation based on the Particle Swarm optimization (PSO) to the obtained path. And make manipulators perform the surgery without collision for a suitable period. Finally, we carry out the system simulation, physical experiments, and dual-arm obstacle avoidance experiments based on the above autonomous obstacle avoidance algorithm. Experiments show that this study can make the dual-arm surgical robot move quickly and smoothly on the premise of avoiding obstacles, which provides a reference for the automation of surgical robots.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Viia-hand: a Reach-and-grasp Restoration system Integrating Voice interaction, Computer vision and Auditory feedback for Blind Amputees

arXiv

引用

arXiv 2023年

作者： Peng, Chunhao Yang, Dapeng Cheng, Ming Dai, Jinghui Zhao, Deyu Jiang, Li State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin150080 China Artificial Intelligence Laboratory Harbin Institute of Technology Harbin150080 China

Visual feedback plays a crucial role in the process of amputation patients completing grasping in the field of prosthesis control. However, for blind and visually impaired (BVI) amputees, the loss of both visual and grasping abilities makes the "easy" reach-and-grasp task a feasible challenge. In this paper, we propose a novel multi-sensory prosthesis system helping BVI amputees with sensing, navigation and grasp operations. It combines modules of voice interaction, environmental perception, grasp guidance, collaborative control, and auditory/tactile feedback. In particular, the voice interaction module receives user instructions and invokes other functional modules according to the instructions. The environmental perception and grasp guidance module obtains environmental information through computer vision, and feedbacks the information to the user through auditory feedback modules (voice prompts and spatial sound sources) and tactile feedback modules (vibration stimulation). The prosthesis collaborative control module obtains the context information of the grasp guidance process and completes the collaborative control of grasp gestures and wrist angles of prosthesis in conjunction with the user's control intention in order to achieve stable grasp of various objects. This paper details a prototyping design (named viia-hand) and presents its preliminary experimental verification on healthy subjects completing specific reach-and-grasp tasks. Our results showed that, with the help of our new design, the subjects were able to achieve a precise reach and reliable grasp of the target objects in a relatively cluttered environment. Additionally, the system is extremely user-friendly, as users can quickly adapt to it with minimal training. Copyright © 2023, The Authors. All rights reserved.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Smart Plant Application of Autonomous Decentralized systems with the Introduction of Edge Nodes 15

Smart Plant Application of Autonomous Decentralized Systems ...

引用

15th International Conference on Advanced Computer Theory and Engineering, ICACTE 2022

作者： Wang, Jingyang Zou, Tao Yang, Zhijia Wang, Hongrui Key Laboratory Of Networked Control System Institutes For Robotics And Intelligent Manufacturing Shenyang Institute Of Automation Chinese Academy Of Sciences Shenyang China Guangzhou University College Of Mechanical And Electrical Engineering Guangzhou China State Key Laboratory Of Robotics Key Laboratory Of Networked Control System Institutes For Robotics And Intelligent Manufacturing Shenyang Institute Of Automation Chinese Academy Of Sciences Shenyang China University Of Chinese Academy Of Sciences Beijing China

ISBN: (数字)9781665481809

ISBN: (纸本)9781665481809

This paper illustrates the application of an autonomous decentralized system introducing edge nodes in a smart factory. The nodes include atomic nodes, the most fundamental autonomic units constituting an autonomous decentralized system that are capable of independently extracting information from the data field and processing that information internally and simultaneously transmitting the processing results and other internal information to the data field proactively in a broadcast mode, where the transmitted information circulates in the data field. The group-based data field enables the sharing of information by atomic nodes. Grouped management nodes are the managers of the whole group, and edge nodes are the terminal intelligent control units of an intelligent autonomous decentralized system. Improvements in system decentralization, fault tolerance characteristics and flexibility of the system can lead to homogenization, autonomous control and autonomous coordination in complex manufacturing environments, as well as excellent online fault tolerance, online expansion and online maintenance of the system. In addition, it enhances the ever-changing and evolving control requirements by deconstructing the complexity in the system. © 2022 IEEE.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Failure Envelopes of Single-Plate Rigid Helical Anchors for Floating Offshore Wind Turbine

引用

China Ocean Engineering 2023年第6期37卷 1000-1010页

作者： CHEN Hong-zhen WANG Le TIAN Ying-hui ZHANG Chun-hui SHEN Zhi-chao LIU Meng-meng State Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation Tianjin UniversityTianjin 300072China State Key Laboratory of Coastal and Offshore Engineering Dalian University of TechnologyDalian 116024China Department of Infrastructure Engineering Faculty of Engineering and Information Technology The University of MelbourneVictoria 3010Australia School of Civil Engineering Hebei University of Science and TechnologyShijiazhuang 050018China Hebei Technological Innovation Center of Disaster Prevention and Mitigation Engineering of Geotechnical and Structural System Shijiazhuang 050018China School of Marine Science and Engineering South China University of TechnologyGuangzhou 511442China Institute of Ocean Energy and Intelligent Construction Tianjin University of TechnologyTianjin 300384China

Helical anchor is a kind of novel foundation for floating offshore wind turbines,which should be subjected to combined tensile loading caused by wind,wave and ***,the research about the capacity of helical anchor was mainly examined under uniaxial loading and scarcely explored under combined *** this study,three-dimensional finite element limit analysis is adopted to assess the bearing capacities of single-plate rigid helical anchors with different ratios of helix to shaft diameter,D_(H)/D_(S) and embedment ratios L/D_(S).Result shows that the vertical,horizontal and moment bearing capacities increase with increasing D_(H)/D_(S) and L/D_(S).The normalized V-H failure envelopes expands with increasing L/D_(S),while the normalized V-M failure envelopes tend to contract with the increase of D_(H)/D_(S).With increasing D_(H)/D_(S) or decreasing L/D_(S),the normalized H-M failure envelopes expand when the horizontal and moment loading act in the same direction and contract when they act in the opposite *** effect of D_(H)/D_(S) and L/D_(S) on the shape of H-M failure envelope become insignificant when L/D_(S)≥4.A series of failure mechanisms under different loading conditions were observed and can be used to explain the ***,a series of approximate expressions were proposed to fit the uniaxial bearing capacities and the failure envelopes.

关键词： helical anchors bearing capacity failure envelope clay

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Signal Demodulation in Underwater Wireless Optical Communications

引用

China Communications 2024年第5期21卷 297-313页

作者： Ma Shuai Yang Lei Ding Wanying Li Hang Zhang Zhongdan Xu Jing Li Zongyan Xu Gang Li Shiyin Department of Strategic and Advanced Interdisciplinary Research Peng Cheng LaboratoryShenzhen 518055China School of Information and Control Engineering China University of Mining and TechnologyXuzhou 221116China Data-driven Intelligent Information System Laboratory Shenzhen Research Institute of Big DataShenzhen 518172China Ocean College Zhejiang UniversityZhoushan 316021China State Key Laboratory of Millimeter Waves Southeast UniversityNanjing 210096China

The underwater wireless optical communication(UWOC)system has gradually become essential to underwater wireless communication *** other existing works on UWOC systems,this paper evaluates the proposed machine learningbased signal demodulation methods through the selfbuilt experimental *** on such a platform,we first construct a real signal dataset with ten modulation ***,we propose a deep belief network(DBN)-based demodulator for feature extraction and multi-class feature *** also design an adaptive boosting(Ada Boost)demodulator as an alternative scheme without feature filtering for multiple modulated ***,it is demonstrated by extensive experimental results that the Ada Boost demodulator significantly outperforms the other *** also reveals that the demodulator accuracy decreases as the modulation order increases for a fixed received optical power.A higher-order modulation may achieve a higher effective transmission rate when the signal-to-noise ratio(SNR)is higher.

关键词： AdaBoost DBN machine learning signal demodulation

来源：评论

学校读者我要写书评

暂无评论

SkeletonMix: A Mixup-Based Data Augmentation Framework for Skeleton-Based Action Recognition

SkeletonMix: A Mixup-Based Data Augmentation Framework for S...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zongye Zhang Huanyu Zhou Qingjie Liu Yunhong Wang State Key Laboratory of Virtual Reality Technology and System Beihang University Beijing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Skeleton-based human action recognition has received widespread attention for its robustness to changes in the background and appearance of actors compared to the RGB modality. Data augmentation is widely used to explicitly regularize the model to prevent overfitting, especially when the number of labeled samples is scarce. However, compared to various augmentation methods available for the RGB modality, there are fewer works on the skeleton modality, especially a model-agnostic augmentation method that can be easily integrated into multiple models. We address the problem by proposing a comprehensive data augmentation framework named SkeletonMix, which contains a pair sample selection module for mixup and random augmentations tailored for skeleton modality. SkeletonMix is a non-learning framework and can be applied to different models in a plug-and-play manner. Extensive experiments on NTU RGB+D, NTU RGB+D 120 and PKU-MMD datasets demonstrate the effectiveness of our proposed framework under limited labeled training data. Our proposed method boosts the performance by a maximum of 7.5% on scarce training data setup (5% of training data).

关键词： Training data Speech recognition Signal processing Data augmentation Skeleton Robustness Data models Human activity recognition Speech processing Overfitting

来源：评论

学校读者我要写书评

暂无评论

Robust Robot Control Policy Learning Based on Gaussian Mixture Variational Autoencoders

Robust Robot Control Policy Learning Based on Gaussian Mixtu...

引用

2024 China Automation Congress, CAC 2024

作者： Dong, Qingwei Zeng, Peng Zang, Chuanzhi Wan, Guangxi Dong, Xiaoting Cui, Shijie State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang110016 China University of Chinese Academy of Sciences Beijing100049 China Shenyang University of Technology Shenyang110870 China

ISBN: (纸本)9798350368604

Reinforcement learning methods have demonstrated advanced capabilities in training neural network controllers for specific tasks. In the industrial domain, where model discrepancies and state disturbances occur, robot control strategies must exhibit adaptability. Existing studies, such as Generative Motor Reflexes (GMR), employ Variational Autoencoders (VAE) to approximate the latent variable space distribution to a standard normal distribution. However, this method, while straightforward, offers limited capability in representing data. The Gaussian Mixture Variational Autoencoder (GMVAE) algorithm leverages a Gaussian Mixture Model (GMM) as the prior distribution for data generation, enhancing the representation of the original state distribution. To increase the robustness of the trained neural network controllers against unknown states, we utilize GMVAE to represent the original state space, subsequently learning the mapping from latent variables to actions. We further accelerate the policy search process by integrating state representation learning with Guided Policy Search (GPS) methods. Our method was evaluated on robotic reaching tasks. Experimental results validate the robustness of our method against state perturbations. © 2024 IEEE.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Modeling and Planar Trajectory Tracking Control of an MJ-AUV 42

Modeling and Planar Trajectory Tracking Control of an MJ-AUV

引用

42nd Chinese Control Conference, CCC 2023

作者： Omar, Zakarya Liu, Ke-Xian Ren, Chao Meng, Qing-Hao Institute of Robotics and Autonomous Systems School of Electrical and Information Engineering Tianjin University Tianjin Key Laboratory of Intelligent Unmanned Swarm Technology and System Tianjin300072 China Sana'a University Faculty of Engineering Sana'a Yemen

ISBN: (纸本)9789887581543

A Multi-Joint Autonomous Underwater Vehicle (MJ-AUV) is designed and equipped with multiple movable segments or joints that allow for greater maneuverability and flexibility in the water. These joints allow the MJ-AUV to change its shape, orientation, and direction of movement, so it can access tight spaces or navigate in complex underwater environments, making it useful for a variety of tasks in the deep sea, such as surveillance, exploration, and maintenance. However, the multiple joint structure of the MJ-AUV complicates the dynamics of the system. Furthermore, the joints' rotation causes misalignment of the cabins, causing deviations in the desired forces and torques. To address these issues, this study focuses on the 2-dimensional (2D) modeling as well as the trajectory tracking controller of the MJ-AUV. In this setting, the proposed controller utilizes a hierarchically structured robust nonlinear back-stepping and sliding-mode approach. Hence, this controller is simulated under various uncertainties and disturbances and the results show that it presents a global finite-time convergence to a bounded neighborhood of the origin for both position and velocity tracking errors, demonstrating its effectiveness and robustness. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Autonomous underwater vehicles

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：