检索结果-内蒙古大学图书馆

IEEE International Symposium on Multimedia

作者： Martin Piccolrovazzi Michael G. Adam Sebastian Eger Marsil Zakour Eckehard Steinbach Department of Computer Engineering Chair of Media Technology Technical University of Munich School of Computation Information and Technology Munich Institute of Robotics and Machine Intelligence (MIRMI) Munich Germany

ISBN: (纸本)9781665471732

Capturing a digital replica of an environment using hand held devices or mobile mapping systems has become increasingly easy in recent years. However, leveraging large amounts of data for various semantic applications is usually impaired by a costly data annotation process. In this paper, we investigate the topic of object recognition in indoor environments without supervision. We approach the problem from a remapping perspective, where we capture RGB images from the same environment at different times and use the naturally occurring changes to identify single objects. In the first step, we create pairs of images from different recordings and generate object candidates using optical flow and an off-the-shelf region proposal algorithm. Then, we use a self-supervised representation learning framework and cluster the extracted objects. We evaluate the performance of several existing clustering methods in an over-clustering setting, since the number of object classes is unknown in an unsupervised setup. Our experimental validation on a real-world dataset shows that the proposed system can successfully recognize objects and pre-annotate a dataset by exploiting a recapturing process.

关键词： Representation learning Clustering methods Semantics Clustering algorithms Self-supervised learning Indoor environment Recording

来源：评论

学校读者我要写书评

暂无评论

Brain Rhythmic Activity and Hemodynamic Response During Wrist Extension Induced by Functional Electrical Stimulation 17

Brain Rhythmic Activity and Hemodynamic Response During Wris...

引用

17th International Convention on Rehabilitation engineering and Assistive Technology, i-CREATe 2024

作者： Mao, He Peng, Yinghu Zheng, Yue Li, Guanglin Li, Hai Xu, Yongan Fang, Peng Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Cas Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Engineering Laboratory of Neural Rehabilitation Technology Shenzhen518055 China Southern University of Science and Technology Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Shenzhen518055 China Southern Medical University Neurorehabilitation Laboratory Department of Rehabilitation Medicine Shenzhen Hospital Shenzhen518101 China Second Affiliated Hospital Institute of Emergency Medicine Zhejiang University School of Medicine Department of Emergency Medicine Hangzhou310009 China

ISBN: (纸本)9798350355154

Functional electrical stimulation (FES) is an effective therapy for motor function restoration after stroke. To investigate brain activation during FES-induced wrist extension, electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) were simultaneously measured for 6 healthy subjects. We found an increase in oxygenated hemoglobin (HbO) concentration, which was accompanied by a decrease in deoxygenated hemoglobin (HbR) concentration in bilateral sensorimotor cortex (SMC) during movement. We also found event-related desynchronization (ERD) of alpha and beta EEG rhythms during movement and event-related synchronization (ERS) after the cessation of the movement. Neurovascular coupling (NC) was evaluated by the correlation coefficient (cc) between alpha ERD and HbO, and a higher cc value was found on the contralateral hemisphere. These findings demonstrate the potential application of EEG-fNIRS for quantifying NC during FES therapy. © 2024 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Interactive RGB Image Segmentation via Depth-modified Click Encoding and Estimated Depth

Interactive RGB Image Segmentation via Depth-modified Click ...

引用

IEEE International Symposium on Multimedia

作者： Furkan Kaynar Adrian Michl Eckehard Steinbach Information and Technology Department of Computer Engineering Chair of Media Technology Technical University of Munich School of Computation Munich Institute of Robotics and Machine Intelligence (MIRMI) Munich Germany

ISBN: (纸本)9781665471732

Interactive image segmentation separates the object of interest in the scene from the background, using the help of human annotations. Click-based interactive segmentation methods typically receive positive and negative clicks from the user indicating the object of interest and the background. These clicks are encoded into click-maps to be processed further by a neural network. Although the depth information is known to improve the image segmentation accuracy, the previous work encodes only the locations of the clicks into the click-maps. We propose two novel click-map generation methods that modify the conventional click-maps using relative depth information. This depth information is estimated from the monocular RGB image. After retraining the baseline interactive segmentation method with our novel click-maps, the segmentation accuracy improved without requiring any additional input or increasing the network size. Experimental evaluations showed that our method yields a better mean segmentation accuracy on the Berkeley and DAVIS datasets than the baseline using conventional click-maps, and a comparable performance on the GrabCut dataset.

关键词： Training Image segmentation Image coding Annotations Neural networks Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Robust Depth Estimation in Foggy Environments Combining RGB Images and mmWave Radar

Robust Depth Estimation in Foggy Environments Combining RGB ...

引用

IEEE International Symposium on Multimedia

作者： Mengchen Xiong Xiao Xu Dong Yang Eckehard Steinbach Department of Computer Engineering Chair of Media Technology Munich Institute of Robotics and Machine Intelligence (MIRMI) School of Computation Information and Technology Technical University of Munich Munich Germany

ISBN: (纸本)9781665471732

In this paper, we propose a robust depth estimation strategy that uses RGB images and mmWave radar data to deal with limited visibility in foggy environments. While the state-of-the-art RGB or LiDAR-based depth estimation works well in scenarios with good visibility, their performance dramatically degrades in the presence of fog. In contrast, mmWave radar sensors are not affected by fog and hence are a promising complement. To leverage this property of mmWave radar, we combine RGB image-based depth estimation with radar information. The proposed combination is an extension of the Sparse-to-Dense (S2D) model. Moreover, a weight-based sensor fusion strategy is presented to improve system performance. Our experiments show that a fog density of meteorological optical range (MOR) less than 50m leads to strongly degraded performance for RGB image-based and LiDAR-based depth estimation. For a MOR of 30m in our dataset, the experiments show an improvement of 26% in mean square error for our proposed approach compared to the combination of RGB images and LiDAR data.

关键词： Meteorological radar Laser radar System performance Estimation Radar imaging Sensor fusion Optical imaging

来源：评论

学校读者我要写书评

暂无评论

Energy-efficient Computation-In-Memory Architecture using Emerging Technologies

Energy-efficient Computation-In-Memory Architecture using Em...

引用

International Conference on Microelectronics, ICM

作者： Rajendra Bishnoi Sumit Diware Anteneh Gebregiorgis Simon Thomann Sara Mannaa Bastien Deveautour Cédric Marchand Alberto Bosio Damien Deleruyelle Ian O’Connor Hussam Amrouch Said Hamdioui Computer Engineering Laboratory Delft University of Technology The Netherlands Chair of AI Processor Design Munich Institute of Robotics and Machine Intelligence Technical University of Munich Germany University Lyon ECL INSA Lyon CPE Lyon Institut de Nanotechnologies de Lyon France

Deep Learning (DL) has recently led to remark-able advancements, however, it faces severe computation related challenges. Existing Von-Neumann-based solutions are dealing with issues such as memory bandwidth limitations and energy inefficiency. Computation-In-Memory (CIM) has the potential to address this problem by integrating processing elements directly into the memory architecture, reducing data movement and enhancing the overall efficiency of the system. In this work, we propose CIM architecture using three distinct emerging technologies. Firstly, a CIM architecture utilizing Ferroelectric Field-Effect Transistors (FeFET) is shown and the resulting errors from the analog compute scheme are injected into the emerging algorithm of Hyperdimensional Computing. Subsequently, we explore Vertical Nanowire Field-Effect Transistors (VNWFETs) based CIM within a 3D computing architecture, demonstrating improved energy efficiency and reconfigurability for CIM. Additionally, we improve the accuracy of the Resistive Random Access Memories (RRAM) based CIM architecture using two mapping-based solutions. These three technologies exhibit non-volatile characteristics, and when integrated into the CIM architecture, they yield significant advantages, including enhanced energy efficiency, reliability, and accuracy in computing processes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MotionTrack: Learning Robust Short-term and Long-term Motions for Multi-Object Tracking

arXiv

引用

arXiv 2023年

作者： Qin, Zheng Zhou, Sanping Wang, Le Duan, Jinghai Hua, Gang Tang, Wei National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China School of Software Engineering Xi’an Jiaotong University China Wormpex AI Research China University of Illinois Chicago United States

The main challenge of Multi-Object Tracking (MOT) lies in maintaining a continuous trajectory for each target. Existing methods often learn reliable motion patterns to match the same target between adjacent frames and discriminative appearance features to re-identify the lost targets after a long period. However, the reliability of motion prediction and the discriminability of appearances can be easily hurt by dense crowds and extreme occlusions in the tracking process. In this paper, we propose a simple yet effective multi-object tracker, i.e., MotionTrack, which learns robust short-term and long-term motions in a unified framework to associate trajectories from a short to long range. For dense crowds, we design a novel Interaction Module to learn interaction-aware motions from short-term trajectories, which can estimate the complex movement of each target. For extreme occlusions, we build a novel Refind Module to learn reliable long-term motions from the target’s history trajectory, which can link the interrupted trajectory with its corresponding detection. Our Interaction Module and Refind Module are embedded in the well-known tracking-by-detection paradigm, which can work in tandem to maintain superior performance. Extensive experimental results on MOT17 and MOT20 datasets demonstrate the superiority of our approach in challenging scenarios, and it achieves state-of-the-art performances at various MOT metrics. Copyright © 2023, The Authors. All rights reserved.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Pedestrian Simulation by Learning from Online Human Demonstration with VR Headset

Pedestrian Simulation by Learning from Online Human Demonstr...

引用

International Conference on Intelligent Transportation

作者： Feiyang Yu Tingting Long Zhichao Cui Linhai Xu Li Li Chi Zhang National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an Shaanxi China Department of Automation BNRist Tsinghua University Beijing China School of Electronics and Control Engineering Chang'an University.

Pedestrian simulation is essential for verifying the safety of autonomous vehicles in simulators. The goal of pedes-trian simulation is to create realistic virtual representations of the pedestrian. However, current simulations lack empirical knowledge about real pedestrian behavior. In this paper, we propose a virtual reality (VR)-based method that enables real-time interaction between an individual and a simulated traffic environment. Our human-in-the-loop approach incorporates the empirical knowledge of a VR user into the simulator. The user can view the traffic simulation by wearing a VR device, which includes a head-mounted display and two controllers and provides sparse sensor data about their position and rotation. We combine representation learning techniques to develop a full-body motion-tracking agent that takes in sparse signals from the VR device and simulates full-body motions. This agent can estimate the full-body pose and visualize a pedestrian avatar in a traffic simulation scene. To validate our method, we constructed a VR traffic simulation environment and demonstrated that our approach can produce a pedestrian avatar whose behavior closely resembles that of the real VR user.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Transcranial Magnetic Stimulation Robotic Assistant: towards a fully automated stimulation session

Transcranial Magnetic Stimulation Robotic Assistant: towards...

引用

IEEE/RAS-EMBS International Conference on Biomedical robotics and Biomechatronics (BioRob)

作者： Ioannis Xygonakis Riccardo Seganfreddo Mazin Hamad Samuel Schneider Axel Schröder Sandro M. Krieg Bernhard Meyer Lorenzo Chiari Melissa Zavaglia Sami Haddadin Technical University of Munich TUM School of Computation Information and Technology Chair of Robotics and Systems Intelligence Munich Institute of Robotics and Machine Intelligence (MIRMI) Germany Department of Electrical Electronic and Information Engineering University of Bologna Italy Department of Neurosurgery Klinikum rechts der Isar Technical University of Munich School of Medicine Germany Department of Neurosurgery Heidelberg University Hospital Germany Centre for Tactile Internet with Human-in-the-Loop (CeTI) Technische Universität Dresden Germany

ISBN: (数字)9798350386523

ISBN: (纸本)9798350386530

Ahstract- Transcranial Magnetic Stimulation (TMS) is a non-invasive method for brain stimulation, commonly employed in depression treatment and pre-operative planning. With an average session duration of 30 minutes, the manual handling and positioning of the TMS coil during the treatment may become strenuous for the operator. Robotized TMS systems can help automate the overall process. Our study introduces the TMS Robotic Assistant (TMS-RA) as an end-to-end system integrated with a commercial navigated TMS. TMS-RA utilizes online calibration, eliminating additional calibration steps needed before each TMS session. Furthermore, it can be adapted for use with mobile-based nTMS systems commonly found in clinics. We have implemented an error-compensation scheme to mitigate positioning errors, effectively constraining both position and orientation errors to final mean values of 1.18 mm and $0.57^{\circ}$ . Comparing our system with a highly experienced medical expert, it is more precise and has significantly lower orientation errors, with a $3^{\circ}$ difference, while the position errors were slightly worse but remained comparable.

关键词： Coils Accuracy Transcranial magnetic stimulation Navigation Biomechatronics Depression Calibration Planning Brain stimulation Robots

来源：评论

学校读者我要写书评

暂无评论

ERetinex: Event Camera Meets Retinex Theory for Low-Light Image Enhancement

arXiv

引用

arXiv 2025年

作者： Guo, Xuejian Tian, Zhiqiang Wang, Yuehang Li, Siqi Jiang, Yu Du, Shaoyi Gao, Yue the School of Software Engineering Xi’an Jiaotong University Xi’an710049 China the College of Computer Science and Technology Jilin University Changchun130012 China the National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an710049 China BNRist THUIBCS KLISS BLBCI School of Software Tsinghua University Beijing100084 China

Low-light image enhancement aims to restore the under-exposure image captured in dark scenarios. Under such scenarios, traditional frame-based cameras may fail to capture the structure and color information due to the exposure time limitation. Event cameras are bio-inspired vision sensors that respond to pixel-wise brightness changes asynchronously. Event cameras’ high dynamic range is pivotal for visual perception in extreme low-light scenarios, surpassing traditional cameras and enabling applications in challenging dark environments. In this paper, inspired by the success of the retinex theory for traditional frame-based low-light image restoration, we introduce the first methods that combine the retinex theory with event cameras and propose a novel retinex-based low-light image restoration framework named ERetinex. Among our contributions, the first is developing a new approach that leverages the high temporal resolution data from event cameras with traditional image information to estimate scene illumination accurately. This method outperforms traditional image-only techniques, especially in low-light environments, by providing more precise lighting information. Additionally, we propose an effective fusion strategy that combines the high dynamic range data from event cameras with the color information of traditional images to enhance image quality. Through this fusion, we can generate clearer and more detail-rich images, maintaining the integrity of visual information even under extreme lighting conditions. The experimental results indicate that our proposed method outperforms state-of-the-art (SOTA) methods, achieving a gain of 1.0613 dB in PSNR while reducing FLOPS by 84.28%. The code is available at https://***/lodew920/ERetinex. © 2025, CC BY.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Spatial and Temporal Analysis of Normal and Shear Forces During Grasping, Manipulation and Social Activities

Spatial and Temporal Analysis of Normal and Shear Forces Dur...

引用

IEEE International Conference on Rehabilitation robotics (ICORR)

作者： Theophil Spiegeler Castañeda Joana Matos Patricia Capsi-Morales Cristina Piazza Department of Computer Engineering School of Computation Information and Technology Technical University of Munich (TUM) Munich Germany Munich Institute of Robotics and Machine Intelligence Technical University of Munich (TUM) Munich Germany Faculty of Engineering University of Porto (U.Porto) Porto Portugal

Extensive research has established and widely acknowledged the important contribution of human hand sensory receptors in providing tactile feedback. The absence of these receptors results in a poor perception of the environment, impairing our efficient manipulation skills. Although the literature emphasizes the significance of normal forces in human grasping, further investigations should point toward the role of shear forces in this process. This paper presents an analysis of human everyday grasping activities through the use of 20 three-axis magnetic soft skin force sensors, in the form of rings and bands, that measure both normal and shear forces. Our study includes twelve tasks that cover various grasping requirements. Results highlight the importance of spatial information and the usefulness of shear forces in the prediction of unexpected changes that can not be always observed in normal forces. Tactile sensing can ultimately be integrated into prosthetic and rehabilitation devices for improved control and potentially provide sensory feedback to the user.

关键词：

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：