检索结果-内蒙古大学图书馆

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Liu, Shuo Jiang, Minglv Lu, Yuanliang Wang, Jianji Zheng, Nanning National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Xi'an710049 China School of Software Engineering Xi'An Jiaotong University Xi'an710049 China

ISBN: (纸本)9798350337020

Assembly inspection is a crucial aspect of smart manufacturing to guarantee the quality of products. To address the challenges posed by complex workshop scenes, an assembly inspection algorithm based on multi-view stereo is proposed. Specifically, to enhance the accuracy of reconstruction, the multi-view stereo method utilizes segmentation attention-assisted depth estimation and point cloud fusion to mitigate the noise in the reconstructed point cloud. And the proposed method introduces normalized depth loss to improve the reconstruction ability of foreground pixels. Futhermore, this paper presents a novel 3D point cloud-based size estimation algorithm capable of accurately estimating the size of small objects in complex working scenes. Experimental results demonstrate the efficacy of the improved multi-view stereo algorithm in assembly scenes and verify the superiority of the proposed size estimation algorithm. © 2023 IEEE.

关键词： assembly inspection multi-view stereo point cloud segmentation attention size estimation

来源：评论

学校读者我要写书评

暂无评论

A genetic algorithm-based method to modulate the difficulty of serious games along consecutive robot-assisted therapy sessions

引用

Computers in Biology and Medicine 2024年 181卷 109033-109033页

作者： Martinez-Pascual, David Catalán, José M. Lledó, Luis D. Blanco-Ivorra, Andrea Vales, Yolanda Garcia-Aracil, Nicolas Robotics and Artificial Intelligence Group of the Bioengineering Institute Miguel Hernández University Avda. de la Universidad Alicante Elche03202 Spain

Background and Objective: One of the biggest challenges during neurorehabilitation therapies is finding an appropriate level of therapy intensity for each patient to ensure the recovery of movement of the affected limbs while maintaining motivation. Different studies have proposed adapting the difficulty of exercises based on psychophysiological state, based on success rate, or by modeling the user's skills. However, all studies propose solutions for a single session, requiring a calibration process before using it in each session. We propose a dynamic adaptation method that can be used during different rehabilitation sessions, without the need for recalibration between sessions. Methods: The adaptation architecture is based on a genetic algorithm that aims to maintain a certain score level and to motivate the user to move. The method has been evaluated with two serious games for five sessions using a rehabilitation robot. A common initial evaluation was made for all the users involved in the study, and the game parameters that best suited each user from the previous session were introduced as the starting point of the next session. In addition, the desired score rate was lowered between sessions to increase the difficulty level. The psychophysiological state of the users was measured based on the Self-Assessment Manikin test, as well as different cardiorespiratory and galvanic skin response signals were analyzed. Results: The adaptation architecture proposed can find those game parameters that maximize the user movement for both games. In one of the games, the score rate set for each session is followed with high fidelity. The degree of personalization in the games increases between sessions as the dispersion of the game parameters grows. The Self-Assessment Manikin test and the physiological signals results would indicate that the psychophysiological state remains equal between sessions despite an increase in game difficulty. Conclusions: The genetic algorithm-bas

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

How Do In-Context Examples Affect Compositional Generalization?

arXiv

引用

arXiv 2023年

作者： An, Shengnan Lin, Zeqi Fu, Qiang Chen, Bei Zheng, Nanning Jian-Guang, L.O.U. Zhang, Dongmei Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China Microsoft Corporation United States

Compositional generalization—understanding unseen combinations of seen primitives—is an essential reasoning capability in human intelligence. The AI community mainly studies this capability by fine-tuning neural networks on lots of training samples, while it is still unclear whether and how in-context learning—the prevailing few-shot paradigm based on large language models—exhibits compositional generalization. In this paper, we present COFE, a test suite to investigate in-context compositional generalization. We find that the compositional generalization performance can be easily affected by the selection of in-context examples, thus raising the research question what the key factors are to make good in-context examples for compositional generalization. We study three potential factors: similarity, diversity and complexity. Our systematic experiments indicate that in-context examples should be structurally similar to the test case, diverse from each other, and individually simple. Furthermore, two strong limitations are observed: in-context compositional generalization on fictional words is much weaker than that on commonly used ones;it is still critical that the in-context examples should cover required linguistic structures, even though the backbone model has been pre-trained on large corpus. We hope our analysis would facilitate the understanding and utilization of in-context learning paradigm. © 2023, CC BY-NC-SA.

关键词： Linguistics

来源：评论

学校读者我要写书评

暂无评论

Functional Connectivity Analysis Revealed Abnormal Brain Network Pattern in Parkinson's Disease with Freezing of Gait 42

Functional Connectivity Analysis Revealed Abnormal Brain Net...

引用

42nd Chinese Control Conference, CCC 2023

作者： Shu, Zhilin Liu, Jinrui Song, Ting Han, Jianda Yu, Ningbo College of Artificial Intelligence Nankai University Tianjin300350 China Nankai University Engineering Research Center of Trusted Behavior Intelligence Ministry of Education Tianjin300350 China Nankai University Tianjin Key Laboratory of Intelligent Robotics Tianjin300350 China Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute of Nankai University Shenzhen518083 China

ISBN: (纸本)9789887581543

Patients with Parkinson's disease (PD) commonly have deficits in lower-limb movements. Freezing of gait (FOG), a type of gait disturbance of PD, is a complex motor symptom exhibited as inability to generate an effective gait. The mechanism of lower-limb dysfunction of PD, especially FOG, remains to be explored. To investigate the neural mechanisms related to FOG in PD, we evaluated 39 subjects (13 PD patients with FOG, 13 PD patients without FOG, and 13 healthy subjects) based on electroencephalogram (EEG) signals while they performed the pedaling task. We conducted functional connectivity analysis using Magnitude Squared Coherence (MSC) and calculated the metrics to investigate the characteristics of the brain network among the three groups. The results indicated that PD patients were associated with reduced functional integration and segregation of brain network, subserving both motor and cognitive abilities. The reduced functional integration and segregation of cognitive processes were correlated with the severity of FOG. This study revealed the differences in brain networks between PD patients and healthy controls and provided insights into the neural mechanism of FOG. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

A Novel Surface Electromyography Signals Denoising Method for Gait Phase Classification 42

A Novel Surface Electromyography Signals Denoising Method fo...

引用

42nd Chinese Control Conference, CCC 2023

作者： Sun, Yubo Guo, Zhenhui Feng, Yuxin Yu, Ningbo College of Artificial Intelligence Nankai University Tianjin300350 China Engineering Research Center of Trusted Behavior Intelligence Ministry of Education Nankai University Tianjin300350 China Nankai University Tianjin Key Laboratory of Intelligent Robotics Tianjin300350 China Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute Nankai University Shenzhen518083 China

ISBN: (纸本)9789887581543

Surface electromyography (sEMG) is a non-invasive technique for recording electrical activity and is extensively employed in gait phase classification. However, the measurement process of the sEMG signal inevitably introduces noise. Consequently, denoising is required for gait analysis from the collected sEMG signal. A novel denoising method ICEEMDAN-DFA-WD has been proposed in this paper. The sEMG signals were decomposed into several intrinsic mode functions (IMF) through an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN). The Hurst exponents were simulated by detrended fluctuation analysis (DFA) to distinguish noisy IMF and desired IMF. During the walking process, the lower limbs' sEMG exhibits strong periodicity, and the desired IMF yields higher Hurst exponents than the noisy IMF. For the noisy IMF, noise removal was performed through further decomposition by wavelet transform, followed by a hard threshold. The processed IMFs were reconstructed to denoise sEMG. In this study, experiments were conducted to compare with the existing denoising methods WD, EMD-IIT, EEMD-IWT, VMD-SIT. The proposed method can obtain better denoising effect and gait phase classification accuracy. The proposed method provides a more effective way for sEMG denoising. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Empirical mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Descriptor: A Comprehensive LiDAR Point Cloud Dataset of Telecommunication Towers (TowerScope)

IEEE Data Descriptions

引用

IEEE Data Descriptions 2024年 1卷 103-108页

作者： Amzar Omairi Zool Hilmi Ismail Mohd Ibrahim Shapiai Malaysia-Japan International Institute of Technology Universiti Teknologi Malaysia Kuala Lumpur Malaysia Center for Artificial Intelligence and Robotics Universiti Teknologi Malaysia Kuala Lumpur Malaysia

This article describes a dataset of light detection and ranging (LiDAR) point clouds captured using the DJI Matrice 300 equipped with the DJI Zenmuse L1 sensor. The TowerScope point cloud dataset contains 29 LiDAR scans from telecommunication towers in Log ASCII Standard (LAS) and wavefront OBJect (OBJ) formats, capturing the main tower or mast structure, antennas, transmission equipment, and supporting infrastructure installed on telecommunication towers from various sites. This dataset is valuable for researchers working in telecommunications, structural analysis, and remote sensing, offering high-resolution data that can be used in various research projects, such as automated structural integrity analysis (load distribution), artificial intelligence, and AI-assisted tower maintenance, and any LiDAR related research. The data are made available to the research community to facilitate further development and analysis in telecommunication tower-related applications. IEEE SOCIETY/COUNCIL Signal Processing Society (SPS) DATA TYPE/LOCATION Point Cloud; Malaysia DATA DOI/PID 10.7910/DVN/MKM9FO

关键词： Laser radar Poles and towers Communications technology Accuracy Drones Three-dimensional displays Maintenance Point cloud compression Solid modeling Predictive models

来源：评论

学校读者我要写书评

暂无评论

D2S: Towards Efficient Sparse 3D Object Detection via Dense to Sparse Knowledge Distillation

D2S: Towards Efficient Sparse 3D Object Detection via Dense ...

引用

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

作者： Yuqi Huang Longjun Liu Yingke Gao Haonan Zhang Haoteng Li Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China Beijing Institute of Control Engineering China Academy of Space Technology Beijing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

LiDAR-based 3D object detection is widely used in high-level autonomous driving schemes. However, the cumbersome modules in most 3D detectors lead to substantial computational overhead. Despite knowledge distillation (KD) is an effective approach for compressing models, previous methods cannot be extended to the dense-to-sparse paradigm. To this end, we propose a simple yet effective Dense to Sparse Knowledge Distillation (D2S) framework for accelerating 3D detectors. Firstly, to compensate for the difference in predicted location between dense and sparse detectors, we introduce a lightweight feature diffusion (FeaD) module for spreading important features. Secondly, to achieve high performance, we propose a dual-stream distillation scheme to transfer knowledge. In this scheme, we align both of the feature and category prediction between distillation pairs at important positions. Extensive experiments on KITTI and Waymo Open Dataset demonstrate the effectiveness of our method. For example, on KITTI dataset, the sparse detector we obtained surpasses VoxelNeXt with around 2.0× fewer parameters and 1.6× fewer FLOPs.

关键词： Three-dimensional displays Computational modeling Detectors Object detection Signal processing Feature extraction Acoustics Speech processing Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Vision mamba: efficient visual representation learning with bidirectional state space model 24

Vision mamba: efficient visual representation learning with ...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Lianghui Zhu Bencheng Liao Qian Zhang Xinlong Wang Wenyu Liu Xinggang Wang School of EIC Huazhong University of Science & Technology Institute of Artificial Intelligence Huazhong University of Science & Technology and School of EIC Huazhong University of Science & Technology Horizon Robotics Beijing Academy of Artificial Intelligence

Recently the state space models (SSMs) with efficient hardware-aware designs, i.e., the Mamba deep learning model, have shown great potential for long sequence modeling. Meanwhile building efficient and generic vision backbones purely upon SSMs is an appealing direction. However, representing visual data is challenging for SSMs due to the position-sensitivity of visual data and the requirement of global context for visual understanding. In this paper, we show that the reliance on self-attention for visual representation learning is not necessary and propose a new generic vision backbone with bidirectional Mamba blocks (Vim), which marks the image sequences with position embeddings and compresses the visual representation with bidirectional state space models. On ImageNet classification, COCO object detection, and ADE20k semantic segmentation tasks, Vim achieves higher performance compared to well-established vision transformers like DeiT, while also demonstrating significantly improved computation & memory efficiency. For example, Vim is 2.8× faster than DeiT and saves 86.8% GPU memory when performing batch inference to extract features on images with a resolution of 1248×1248. The results demonstrate that Vim is capable of overcoming the computation & memory constraints on performing Transformer-style understanding for high-resolution images and it has great potential to be the next-generation backbone for vision foundation models. Code and models are released at https://***/hustvl/Vim

关键词：

来源：评论

学校读者我要写书评

暂无评论

SurgPLAN: Surgical Phase Localization Network for Phase Recognition

SurgPLAN: Surgical Phase Localization Network for Phase Reco...

引用

IEEE International Symposium on Biomedical Imaging

作者： Xingjian Luo You Pang Zhen Chen Jinlin Wu Zongmin Zhang Zhen Lei Hongbin Liu Centre for Artificial Intelligence and Robotics HKISI CAS Hong Kong SAR China MAIS Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China

ISBN: (数字)9798350313338

ISBN: (纸本)9798350313345

Surgical phase recognition is crucial to providing surgery understanding in smart operating rooms. Despite great progress in automatic surgical phase recognition, most existing methods are still restricted by two problems. First, these methods cannot capture discriminative visual features for each frame and motion information with simple 2D networks. Second, the frame-by-frame recognition paradigm degrades the performance due to unstable predictions within each phase, termed as phase shaking. To address these two challenges, we propose a Surgical Phase LocAlization Network, named SurgPLAN, to facilitate a more accurate and stable surgical phase recognition with the principle of temporal detection. Specifically, we first devise a Pyramid SlowFast (PSF) architecture to serve as the visual backbone to capture multi-scale spatial and temporal features by two branches with different frame sampling rates. Moreover, we propose a Temporal Phase Localization (TPL) module to generate the phase prediction based on temporal region proposals, which ensures accurate and consistent predictions within each surgical phase. Extensive experiments confirm the significant advantages of our SurgPLAN over frame-by-frame approaches in terms of both accuracy and stability.

关键词： Location awareness Visualization Image segmentation Accuracy Image recognition Surgery Stability analysis

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Performance Optimization of a Novel Anthropomorphic Dexterous Hand

引用

IEEE robotics and Automation Letters 2025年第7期10卷 6816-6823页

作者： Ma, Yifan Shang, Weiwei Zhang, Fei Pang, Shunxiang Dai, Shuqing Yi, Zhengkun Li, Guotao Cong, Shuang University of Science and Technology of China Department of Automation Jinzhai Road 96 Hefei 230027 China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Provincial Key Laboratory of Robotics and Intelligent System Shenzhen 518055 China Institute of Automation Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Beijing 100190 China

Grasp and in-hand manipulation of dexterous hands are essential for daily operational tasks. In this study, an optimized design method for a novel anthropomorphic dexterous hand is proposed, which improves the comprehensive performance of grasp and manipulability of the dexterous hand. The constraints of multi-fingered contact force and kinematics are established with full consideration of the applicability to the objects of different sizes and the shape characteristics of finger contact surface. The performance index of in-hand manipulation ability is proposed, and the optimization problem is constructed through the technique for order preference by similarity to ideal solution (TOPSIS) to achieve multi-objective optimization of precision grasp and in-hand manipulation. Finally, simulation and experiment verify the feasibility of our prototype and optimization method. The results show that the in-hand translation space and in-hand rotation range are increased by 20% and 8.5%, respectively. Our dexterous hand has excellent abilities of multi-fingered interaction and in-hand manipulation, which can perform 33 grasp gestures and two/three-fingered in-hand manipulation. This study provides a valuable reference for the design of dexterous hands with anthropomorphic functions. © 2016 IEEE.

关键词： design optimization Dexterous manipulation multifingered hands precision grasp

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：