检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Cong, Runmin Sun, Mengyao Zhang, Sanyi Zhou, Xiaofei Zhang, Wei Zhao, Yao School of Control Science and Engineering Key Laboratory of Machine Intelligence and System Control Ministry of Education Shandong University Jinan Shandong China School of Automation Hangzhou Dianzi University Zhejiang Hangzhou China Institute of Information Science Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing Jiaotong University Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China

Camouflaged object detection (COD) aims to accurately detect objects hidden in the surrounding environment. However, the existing COD methods mainly locate camouflaged objects in the RGB domain, their performance has not been fully exploited in many challenging scenarios. Considering that the features of the camouflaged object and the background are more discriminative in the frequency domain, we propose a novel learnable and separable frequency perception mechanism driven by the semantic hierarchy in the frequency domain. Our entire network adopts a two-stage model, including a frequency-guided coarse localization stage and a detail-preserving fine localization stage. With the multi-level features extracted by the backbone, we design a flexible frequency perception module based on octave convolution for coarse positioning. Then, we design the correction fusion module to step-by-step integrate the high-level features through the prior-guided correction and cross-layer feature channel association, and finally combine them with the shallow features to achieve the detailed correction of the camouflaged objects. Compared with the currently existing models, our proposed method achieves competitive performance in three popular benchmark datasets both qualitatively and quantitatively. The code will be released at https://***/rmcong/FPNet_ACMMM23. © 2023, CC BY-NC-SA.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

RDNet: A reinforced-deformable convolutional network for crowd counting

RDNet: A reinforced-deformable convolutional network for cro...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Qingzhen Shang Yidi Zhuo Jiahui Zhang Jinfu Yang Control engineering from Beijing University of Technology Beijing China Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing China

ISBN: (纸本)9781665481106

Crowd counting is still a challenging task in crowded scenario due to heavy occlusions, appearance variations and perspective distortions. Thanks to the development of deep convolutional neural networks, especially the deformable convo-lution neurral networks, great progress has been made for crowd counting. However, existing deformable convolution networks cannot get satisfactory performance due to limited receptive field that is not conducive to crowded object detection. In this paper, a novel end-to-end Reinforced-Deformable convolutional neural Network (RDNet) is proposed. First, a multi-scale context information fusion strategy with skipping connection and shortcut connection is utilized to obtain rich semantic features and local details of the crowd distribution. Then a reinforced-deformable convolution module is designed as the back-end of the network, which can sample with adaptive offsets and scalars to enlarge the receptive field for more accurate edge information and crowd distribution localization. Finally, an impulse function convolved with cuphead Gaussian kernel is applied to generate density map for counting. Extensive experimental results on four popular datasets, namely, Shangha-itech, UCF _CC_50, WorldEXPO'10, and UCF-QNRF, show that our proposed approach outperforms other state-of-the-art methods.

关键词： Location awareness Convolution Sociology Semantics Object detection Convolutional neural networks Task analysis

来源：评论

学校读者我要写书评

暂无评论

Self-supervised Multi-frame Monocular Depth Estimation with Pseudo-LiDAR Pose Enhancement

Self-supervised Multi-frame Monocular Depth Estimation with ...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Wenhua Wu Guangming Wang Jiquan Zhong Hesheng Wang Zhe Liu MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of Intelligent Control and Management Insititute of Medical Robotics Shanghai Jiao Tong University Shanghai China

Depth estimation is one of the most important tasks in scene understanding. In the existing joint self-supervised learning approaches of depth-pose estimation, depth estimation and pose estimation networks are independent of each other. They only use the adjacent image frames for pose estimation and lack the use of the estimated geometric information. To enhance the depth-pose association, we propose a monocular multi-frame unsupervised depth estimation framework, named PLPE-Depth. There are a depth estimation network and two pose estimation networks with image input and pseudo-LiDAR input. The main idea of our approach is to use the pseudo-LiDAR reconstructed from the depth map to estimate the pose of adjacent frames. We propose depth re-estimation with a better pose between the image pose and the pseudo-LiDAR pose to improve the accuracy of estimation. Besides, we improve the reconstruction loss and design a pseudo-LiDAR pose enhancement loss to facilitate the joint learning. Our approach enhances the use of the estimated depth information and strengthens the coupling between depth estimation and pose estimation. Experiments on the KITTI dataset show that our depth estimation achieves state-of-the-art performance at low resolution. Our source codes will be released on https://***/IRMVLabIPLPE-Depth.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Passivation of degradation path enables high performance perovskite nanoplatelet lasers with high operational stability

引用

Photonics Research 2022年第6期10卷 1440-1452页

作者： GUOHUI LI HUIHUI PI YANFU WEI BOLIN ZHOU YA GAO RONG WEN YUYING HAO HAN ZHANG BENG S.ONG YANXIA CUI College of Physics and Optoelectronics Key Laboratory of Interface Science and Engineering in Advanced MaterialsKey Laboratory of Advanced Transducers and Intelligent Control System of Ministry of EducationTaiyuan University of TechnologyTaiyuan 030024China Collaborative Innovation Centre for Optoelectronic Science and Technology Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong ProvinceCollege of Physics and Optoelectronic EngineeringShenzhen Key Laboratory of Micro-Nano Photonic Information TechnologyGuangdong Laboratory of Artificial Intelligence and Digital Economy(SZ)Shenzhen UniversityShenzhen 518060China Department of Chemistry Research Centre of Excellence for Organic ElectronicsInstitute of Advanced MaterialsHong Kong Baptist UniversityKowloon TongHong KongChina

MAPbI_(3)perovskite has attracted widespread interests for developing low-cost near infrared semiconductor gain ***,it faces the instability issue under operation conditions,which remains a critical *** is found that the instability of the MAPbI_(3)nanoplatelet laser comes from the thermal-induced degradation progressing from the surface defects towards neighboring *** using PbI_(2) passivation,the defect-initiated degradation is significantly suppressed and the nanoplatelet degrades in a layer-by-layer way,enabling the MAPbI_(3)laser to sustain for 4500 s(2.7×10^(7) pulses),which is nearly three times longer than that of the nanoplatelet laser without ***,the PbI_(2) passivated MAPbI_(3)nanoplatelet laser with the nanoplatelet cavity displays a maximum quality factor up to∼7800,the highest reported for all MAPbI_(3)nanoplatelet ***,a high stability MAPbI_(3)nanoplatelet laser that can last for 8500 s(5.1×10^(7) pulses)is demonstrated based on a dual passivation strategy,by retarding the defect-initiated degradation and surface-initiated degradation *** work provides in-depth insights for understanding the operating degradation of perovskite lasers,and the dual passivation strategy paves the way for developing high stability near infrared semiconductor laser media.

关键词： stability initiated high

来源：评论

学校读者我要写书评

暂无评论

Real-time Terrain Recognition based on Transformer and Gait Prediction based on CNN-LSTM-Attention for Exosuit

Real-time Terrain Recognition based on Transformer and Gait ...

引用

2023 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2023

作者： Yang, Fangliang Chen, Chunjie Wang, Zhuo Xu, Yinsheng Liu, Yao Wu, Xinyu University of Chinese Academy of Sciences Shenzhen College of Advanced Technology Shenzhen518055 China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen518055 China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen518055 China Guilin University of Technology Guilin541004 China

ISBN: (纸本)9798350327182

For wearable soft exosuits, an imprecise control strategy can easily injure the wearer, while real-time terrain recognition and accurate gait phase estimation can effectively improve the control strategy of the exosuit and make the wearer comfortable and safe. In this paper, a real-time terrain recognition algorithm based on Transformer and a gait estimation algorithm based on CNN-LSTM-Attention are implemented. The Transformer-based recognition of different terrains improves the recognition accuracy to some extent, and the CNN-LSTM-Attention based feature extraction for temporal signals such as gait phase is also extremely noticeable. Experiments show that the Transformer-based algorithm achieves 99.64 % recognition accuracy in six different terrain environments. In the gait phase estimation experiment, CNN-LSTM-CBAM achieved the best performance with an evaluation index r2 of 0.9221. The aforementioned terrain recognition algorithms and gait phase estimation algorithms may have a positive impact on soft exosuit and dynamic prosthetics research. © 2023 IEEE.

关键词： Landforms

来源：评论

学校读者我要写书评

暂无评论

Diving into underwater: segment anything model guided underwater salient instance segmentation and a large-scale dataset 24

Diving into underwater: segment anything model guided underw...

引用

Proceedings of the 41st International Conference on machine Learning

作者： Shijie Lian Ziyi Zhang Hua Li Wenjie Li Laurence Tianruo Yang Sam Kwong Runmin Cong Hainan University The Hong Kong University of Science and Technology (Guangzhou) Hainan University and Key Laboratory of New Generation Artificial Intelligence Technology & Its Interdisciplinary Applications (Southeast University) Ministry of Education China Hainan University and Huazhong University of Science and Technology and St. Francis Xavier University LIngnan University School of data science Hong Kong Shandong University School of Control Science and Engineering and Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China

With the breakthrough of large models, Segment Anything Model (SAM) and its extensions have been attempted to apply in diverse tasks of computer vision. Underwater salient instance segmentation is a foundational and vital step for various underwater vision tasks, which often suffer from low segmentation accuracy due to the complex underwater circumstances and the adaptive ability of models. Moreover, the lack of large-scale datasets with pixel-level salient instance annotations has impeded the development of machine learning techniques in this field. To address these issues, we construct the first large-scale underwater salient instance segmentation dataset (USIS10K), which contains 10,632 underwater images with pixel-level annotations in 7 categories from various underwater scenes. Then, we propose an Underwater Salient Instance Segmentation architecture based on Segment Anything Model (USIS-SAM) specifically for the underwater domain. We devise an Underwater Adaptive Visual Transformer (UA-ViT) encoder to incorporate underwater domain visual prompts into the segmentation network. We further design an out-of-the-box underwater Salient Feature Prompter Generator (SFPG) to automatically generate salient prompters instead of explicitly providing foreground points or boxes as prompts in SAM. Comprehensive experimental results show that our USIS-SAM method can achieve superior performance on USIS10K datasets compared to the state-of-the-art methods. Datasets and codes are released on Github.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Cross-Layer Patch Alignment and Intra-and-Inter Patch Relations for Knowledge Distillation

Cross-Layer Patch Alignment and Intra-and-Inter Patch Relati...

引用

IEEE International Conference on Image Processing

作者： Yi Zhang Yingke Gao Haonan Zhang Xinyu Lei Longjun Liu National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Beijing Institute of Control Engineering China Academy of Space Technology

To date, most existing distillation methods exploit coarse-grained instance-level information as valuable knowledge to transfer, such as instance logits, instance features and instance relations. However, the fine-grained knowledge of internal regions and relationships between semantic entities within a single instance are overlooked and not fully explored. To address above limitations, we propose a novel fine-grained patch-level distillation method, dubbed as Patch Aware Knowledge Distillation. PAKD rethink knowledge distillation from a new perspective regarding the significance of cross-layer patch alignment and patch relations within and across instances. Specifically, we first devise a novel cross-layer architecture to fuse patches across stages, which is capable of utilizing multi-level information of the teacher to guide one-level learning of the student. Then, we propose cross-layer patch alignment, allowing the student to be aware of patches discriminatively and find the best way to learn from the teacher. Besides, patch relations within and across instances are leveraged to supervise the structural knowledge distillation in the manifold space. We apply our method to image classification and object detection tasks. Consistent improvements over state-of-the-art approaches on different datasets and diverse teacher-student combinations manifest the great potential of our proposed PAKD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Sliding Mode Terminal Guidance Method

An Adaptive Sliding Mode Terminal Guidance Method

引用

Chinese Automation Congress (CAC)

作者： Yichuan Zhu Na Yao Hu Fan National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing Aerospace Automatic Control Institute Beijing China System simulation laboratory Beijing Aerospace Automatic Control Institute Beijing China The third representative office of air force equipment in Beijing area Beijing China

This article designs a sliding mode terminal guidance law for high-speed targets by modifying the approach law. Firstly, the relative motion equation of the chaser and escaper is established; Then adjust the aircraft based on the changes in the distance of the chaser and escaper and approach speed, design a self adaptive approach law, and derive the guidance law. At the same time, prove the stability of the guidance system by constructing a Lyapunov function; Finally, simulation verification was conducted. The simulation results showed that the designed guidance law can achieve the encounter between the chaser and the escaper. By comparing it with proportional guidance under different operating conditions, it was verified that the method has higher adaptability to different operating conditions, thus verifying the effectiveness of the guidance law designed in this paper and providing certain reference value for further research.

关键词：

来源：评论

学校读者我要写书评

暂无评论

About Step-Length of ZeaD（Zhang et al Discretization） Formula 4IgSY for Future Minimization via Fan Equations

About Step-Length of ZeaD（Zhang et al Discretization） Form...

引用

第30届中国控制与决策会议

作者： Yunong Zhang Zhiyuan Qi Huihui Gong Jian Li Binbin Qiu School of Information Science and Technology Sun Yat-sen University(SYSU) Key Laboratory of Autonomous Systems and Networked Control Ministry of Education SYSU-CMU Shunde International Joint Research Institute Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education

Future minimization,i.e.,discrete time-varying minimization,is a difficult and meaningful *** has been successfully solved by Zhang et al using zeroing dynamics(ZD) and discretization *** this paper,a type of discrete-time ZD(DT-ZD) model,which is obtained via utilizing Zea D(Zhang et al Discretization) formula 4 I2)S Y,is analyzed and investigated to ensure its ***,via theoretical guarantees,we propose the step-length domain,or say,the effective domain of the step-length,which makes the discrete-time model ***,we make further efforts to obtain the step-length optimum which provides the optimal stability of the DT-ZD ***,numerical experiments are performed to validate the step-length domain and the step-length optimum of the DT-ZD model for future minimization.

关键词： Future minimization Zeroing dynamics(ZD) Zhang et al discretization(ZeaD) formula Step-length domain Optimum

来源：评论

学校读者我要写书评

暂无评论

Jerk-Level Cyclic Motion Planning and control for Constrained Redundant Robot Manipulators Using Zhang Dynamics:Theoretics

Jerk-Level Cyclic Motion Planning and Control for Constraine...

引用

第30届中国控制与决策会议

作者： Min Yang Yunong Zhang Huanchang Huang Dechao Chen Jian Li School of Information Science and Technology Sun Yat-sen University(SYSU) Key Laboratory of Autonomous Systems and Networked Control Ministry of Education SYSU-CMU Shunde International Joint Research Institute Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education

In this paper,based on Zhang dynamics method（ZDM）,a cyclic motion planning（CMP） scheme is developed to remedy the joint drift phenomenon of redundant robot manipulators constrained by joint physical *** existing works resolved at velocity level or acceleration level,the scheme in this paper is resolved at jerk *** achieve superior performance,the jerk-level CMP scheme is improved by the feedback control,which further leads to a jerk-level cyclic motion planning and control（CMPC） *** cyclic motion criterion and the avoidance of joint physical limits（e.g.,joint-angle,joint-velocity,joint-acceleration and joint-jerk limits） are both considered into the jerklevel CMP and CMPC *** importantly,the correctness of the presented jerk-level schemes is guaranteed by the corresponding engineering-type theoretical analyses and ***,the jerk-level CMPC scheme is reformulated as a dynamical quadratic program（DQP） which can be solved by a piecewise-linear projection equation neural network（PLPENN）,and also can be solved by a suitable numerical algorithm.

关键词： Constrained robot manipulators Joint drift phenomenon Jerk level Feedback control Theoretics

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：