检索结果-内蒙古大学图书馆

Cleo: Smart Glasses to Monitor Consumption of Alcohol and Cigarettes

SN computer Science 2023年第1期4卷 53页

作者： Sethuraman, Sibi Chakkaravarthy Kompally, Pranav Mohanty, Saraju P. Raju, Bksp Kumar Cruz, Meenalosini Vimal School of Computer Science and Engineering and Center of Excellence in Artificial Intelligence and Robotics (AIR) VIT-AP University Amaravati India Department of Computer Science and Engineering University of North Texas Denton 76207 TX United States Department of Information Technology Allen E. Paulson College of Engineering and Computing Georgia Southern University Statesboro GA United States

It is estimated that over 60% of people around the globe consume alcohol and cigars daily. Many people use them beyond the permitted limit, which causes lung cancer, liver and kidney failure. If there is a system that could monitor their intake level, it will alert them in case of excess consumption, which could help them control the intake. To help the users monitor their consumption, we introduce Cleo Eyeglasses in this paper. Cleo is a wearable spectacle device with a mounted camera and single onboard computer that performs custom trained object recognition to identify alcoholic beverages and cigarettes. Upon recognition, a log is automatically maintained in the corresponding mobile application. If the intake exceeds the user threshold, then our system generates an alert to both the user and concerned medical personnel. Cigarette/alcohol consumption can immediately affect the body without noticeable symptoms. Cleo addresses this with keen monitoring of vital body parameters and generates an alert when abnormalities are detected. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Alcohol monitoring CNN computer vision Deep learning Smart glasses Smoke monitoring

来源：评论

学校读者我要写书评

暂无评论

Hard Gate Knowledge Distillation - Leverage Calibration for a Robust and Reliable Language Model

Hard Gate Knowledge Distillation - Leverage Calibration for ...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Lee, Dongkyu Tian, Zhiliang Zhao, Yingxiu Cheung, Ka Chun Zhang, Nevin L. Department of Computer Science and Engineering HKUST Hong Kong College of Computer National University of Defense Technology China NVIDIA AI Technology Center NVIDIA

In knowledge distillation, a student model is trained with supervisions from both knowledge from a teacher and observations drawn from a training data distribution. Knowledge of a teacher is considered a subject that holds inter-class relations which send a meaningful supervision to a student;hence, much effort has been put to find such knowledge to be distilled. In this paper, we explore a question that has been given little attention: "when to distill such knowledge." The question is answered in our work with the concept of model calibration;we view a teacher model not only as a source of knowledge but also as a gauge to detect miscalibration of a student. This simple and yet novel view leads to a hard gate knowledge distillation scheme that switches between learning from a teacher model and training data. We verify the gating mechanism in the context of natural language generation at both the token-level and the sentence-level. Empirical comparisons with strong baselines show that hard gate knowledge distillation not only improves model generalization, but also significantly lowers model calibration error. © 2022 Association for Computational Linguistics.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

A TB-CSP Motor Imagery EEG Recognition Method for Neural-controlled Manipulator System 12

A TB-CSP Motor Imagery EEG Recognition Method for Neural-con...

引用

12th International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2022

作者： Mu, Wei Wang, Pengchao Wang, Junkongshuai Fang, Tao Wang, aiping Niu, Lan Bin, Jianxiong Liu, Lusheng Zhang, Jing Jia, Jie Zhang, Lihua Kang, Xiaoyang Laboratory for Neural Interface and Brain Computer Interface Engineering Research Center of Ai and Robotics Ministry of Education Shanghai Engineering Research Center of Ai and Robotics Moe Frontiers Center for Brain Science State Key Laboratory of Medical Neurobiology Institute of Ai and Robotics Academy for Engineering and Technology Fudan University Shanghai200433 China Ji Hua Laboratory Guangdong Province Foshan528200 China Department of Rehabilitation Medicine Huashan Hospital Fudan University Shanghai200040 China Yiwu Research Institute of Fudan University Zhejiang Yiwu City322000 China Research Center for Intelligent Sensing Zhejiang Lab Hangzhou311121 China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

To develop a manipulator system based on spontaneous EEG signal control, this paper proposes a common spatial pattern method combined with the Bhattacharyya distance of time-frequency signal (TB-CSP) for EEG signal feature extraction. Firstly, for different subjects, the original EEG signal is divided by a specific time window and different frequency band;Secondly, the average Bhattacharyya distance of different frequency bands in a specific time window is calculated to select the optimal time-frequency signal and input it to CSP for feature extraction;Finally, the extracted features are input into SVM and LDA to classify and complete the control of the manipulator. The proposed method achieves 82.6% accuracy on the public dataset BCI IVa and 79.48% accuracy on the self-collected dataset. © 2022 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Real-Time Motion Generation and Data Augmentation for Grasping Moving Objects with Dynamic Speed and Position Changes

Real-Time Motion Generation and Data Augmentation for Graspi...

引用

IEEE/SICE International Symposium on System Integration

作者： Kenjiro Yamamoto Hiroshi Ito Hideyuki Ichiwara Hiroki Mori Tetsuya Ogata Robotics Research Department Connective Automation Innovation Center R&D Group Hitachi Ltd. Institute for AI and Robotics Future Robotics Organization Faculty of Science and Engineering Waseda University

While deep learning enables real robots to perform complex tasks had been difficult to implement in the past, the challenge is the enormous amount of trial-and-error and motion teaching in a real environment. The manipulation of moving objects, due to their dynamic properties, requires learning a wide range of factors such as the object's position, movement speed, and grasping timing. We propose a data augmentation method for enabling a robot to grasp moving objects with different speeds and grasping timings at low cost. Specifically, the robot is taught to grasp an object moving at low speed using teleoperation, and multiple data with different speeds and grasping timings are generated by down-sampling and padding the robot sensor data in the time-series direction. By learning multiple sensor data in a time series, the robot can generate motions while adjusting the grasping timing for unlearned movement speeds and sudden speed changes. We have shown using a real robot that this data augmentation method facilitates learning the relationship between object position and velocity and enables the robot to perform robust grasping motions for unlearned positions and objects with dynamically changing positions and velocities.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Object Discovery by Low-Dimensional Object Motion

arXiv

引用

arXiv 2023年

作者： Safadoust, Sadra Güney, Fatma KUIS AI Center Department of Computer Engineering Koç University Turkey

Recent work in unsupervised multi-object segmentation shows impressive results by predicting motion from a single image despite the inherent ambiguity in predicting motion without the next image. On the other hand, the set of possible motions for an image can be constrained to a low-dimensional space by considering the scene structure and moving objects in it. We propose to model pixel-wise geometry and object motion to remove ambiguity in reconstructing flow from a single image. Specifically, we divide the image into coherently moving regions and use depth to construct flow bases that best explain the observed flow in each region. We achieve state-of-the-art results in unsupervised multi-object segmentation on synthetic and real-world datasets by modeling the scene structure and object motion. Our evaluation of the predicted depth maps shows reliable performance in monocular depth estimation. © 2023, CC BY.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Trajectory Optimization Algorithm of Trajectory Rehabilitation Training Mode for Rehabilitation Robot

Trajectory Optimization Algorithm of Trajectory Rehabilitati...

引用

2022 IEEE International Conference on robotics and Biomimetics, ROBIO 2022

作者： Cen, Yuhui Yuan, Jianjun Ma, Shugen Luo, Jingjing Wang, Hongbo Human Phenome Institute Fudan University Shanghai200438 China The Shanghai Robotics Institute Shanghai University 99 Shangda RD Baoshan District Shanghai China The Department of Robotics Faculty of Science and Engineering Ritsumeikan University 1-1-1 Nojihigashi Shiga Kusatsu Japan Institute of AI and Robotics Academy for Engineering and Technology Fudan University Engineering Research Center of AI & Robotics Ministry of Education Shanghai210043 China

ISBN: (纸本)9781665481090

It is a challenge for robot-assisted rehabilitation therapy to develop a training program with both customized and optimized characteristics. To optimize the scheme of the trajectory rehabilitation training mode, we design a trajectory optimization algorithm with an intelligent decision mechanism. A trajectory smoothing improvement algorithm is proposed to make the trajectory more suitable for rehabilitation training. In addition, we propose an improved genetic algorithm that reduces the optimization time and obtains a better solution. An adaptive variable time weight improved trapezoidal velocity algorithm is innovatively designed by fusing the above two algorithms with an optimal rehabilitation training program as a guide. Based on the experimental platform of the upper limb rehabilitation robot, it is demonstrated that the algorithm can provide comfortable and high-quality rehabilitation training for patients by adaptively optimizing the parameters of the trajectory model according to the training constraints. By combining an intelligent optimization algorithm with a rehabilitation robot trajectory planning algorithm, this paper realizes the unification of customization and optimization of a trajectory rehabilitation training program. © 2022 IEEE.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

LEMMA: Learning Language-Conditioned Multi-Robot Manipulation

引用

IEEE robotics and Automation Letters 2023年第10期8卷 6835-6842页

作者： Gong, Ran Gao, Xiaofeng Gao, Qiaozi Shakiah, Suhaila Thattai, Govind Sukhatme, Gaurav S. UCLA Center for Vision Cognition Learning and Autonomy Los AngelesCA90025 United States Amazon Alexa AI San JoseCA95129 United States USC Viterbi School of Engineering Department of Computer Science Los AngelesCA90089 United States

Complex manipulation tasks often require robots with complementary capabilities to collaborate. We introduce a benchmark for LanguagE-Conditioned Multi-robot MAnipulation (LEMMA) focused on task allocation and long-horizon object manipulation based on human language instructions in a tabletop setting. LEMMA features 8 types of procedurally generated tasks with varying degree of complexity, some of which require the robots to use tools and pass tools to each other. For each task, we provide 800 expert demonstrations and human instructions for training and evaluations. LEMMA poses greater challenges compared to existing benchmarks, as it requires the system to identify each manipulator's limitations and assign sub-tasks accordingly while also handling strong temporal dependencies in each task. To address these challenges, we propose a modular hierarchical planning approach as a baseline. Our results highlight the potential of LEMMA for developing future language-conditioned multi-robot systems. © 2016 IEEE.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral anomaly detection:a performance comparison of existing techniques

引用

International Journal of Digital Earth 2022年第1期15卷 2078-2125页

作者： Noman Raza Shah Abdur Rahman M.Maud Farrukh Aziz Bhatti Muhammad Khizer Ali Khurram Khurshid Moazam Maqsood Muhammad Amin iVISION Lab Electrical Engineering DepartmentInstitute of Space TechnologyIslamabadPakistan AI Enabling Technologies Lab Center of Excellence in Artificial IntelligenceBahria UniversityIslamabadPakistan Department of Electrical and Computer Engineering Pak-Austria Fachhochschule:Institute of Applied Sciences and TechnologyHaripurPakistan Avionics Engineering Department Institute of Space TechnologyIslamabadPakistan

Anomaly detection in Hyperspectral Imagery(HSI)has received considerable attention because of its potential application in several *** anomaly detection algorithms for HSI have been proposed in the literature;however,due to the use of different datasets in previous studies,an extensive performance comparison of these algorithms is *** this paper,an overview of the current state of research in hyperspectral anomaly detection is presented by broadly dividing all the previously proposed algorithms into eight different *** addition,this paper presents the most comprehensive comparative analysis to-date in hyperspectral anomaly detection by evaluating 22 algorithms on 17 different publicly available *** indicate that attribute and edge-preserving filtering-based detection(AED),local summation anomaly detection based on collaborative representation and inverse distance weight(LSAD-CR-IDW)and local summation unsupervised nearest regularized subspace with an outlier removal anomaly detector(LSUNRSORAD)perform better as indicated by the mean and median values of area under the receiver operating characteristic(ROC)***,this paper studies the effect of various dimensionality reduction techniques on anomaly *** indicate that reducing the number of components to around 20 improves the performance;however,any further decrease deteriorates the performance.

关键词： Anomaly detection algorithms hyperspectral imagery deep learning dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Mobile Application System for Online Surveys and Questionnaires

Mobile Application System for Online Surveys and Questionnai...

引用

Mechatronics, IoT and Industrial Informatics (ICMIII), International Conference on

作者： Ramiz Salama Fadi Al-Turjman Chadi Altrjman Department of Computer Engineering AI and Robotics Institute Research Center for AI and IoT Near East University Nicosia Mersin 10 Turkey Artificial Intelligence Engineering Dept AI and Robotics Institute Near East University Nicosia Mersin 10 Turkey Research Center for AI and IoT Faculty of Engineering University of Kyrenia Kyrenia Mersin 10 Turkey Waterloo University canada

The Covid-19 pandemic, which shook the entire world and altered the dynamics of humanity, resulted in drastic alterations in our life. With the pandemic, human history has understood that the digital world is more important than ever. Existing efforts to digitize pre-pandemic surveys have not entirely replaced face-to-face research. The fundamental issue with online survey platforms can be divided into two categories. It is mostly due to the fact that the particular question types used in practically all research cannot be customized and adjusted using a simple internet program, and certain limited alterations cannot be matched to the research patterns. Second, in addition to the inadequacies of these applications, the online services provided are mainly fee-based programs or applications. The application that is proposed to be established and developed in this context aims to overcome the bottleneck experienced by the end user and to supply the users with high-quality tools that they may use in the global world.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reactive Multi-agent Coordination using Auction-based Task Allocation and Behavior Trees

Reactive Multi-agent Coordination using Auction-based Task A...

引用

Control Technology and Applications (CCTA),

作者： Niklas Dahlquist Björn Lindqvist Akshit Saradagi George Nikolakopoulos Department of Computer Science Electrical and Space Engineering Robotics and AI Group Luleå University of Technology Luleå Sweden

This article presents an architecture for multi-agent task allocation and task execution, through the unification of a market-inspired task-auctioning system with Behavior Trees for managing and executing lower level behaviors. We consider the scenario with multi-stage tasks, such as ’pick and place’, whose arrival times are not known a priori. In such a scenario, a coordinating architecture is expected to be reactive to newly arrived tasks and the resulting rerouting of agents should be dependent on the stage of completion of their current multi-stage tasks. In the novel architecture proposed in this article, a central auctioning system gathers bids (cost-estimates for completing currently available tasks) from all agents, and solves a combinatorial problem to optimally assign tasks to agents. For every agent, it’s participation in the auctioning system and execution of an assigned multi-stage task is managed using behavior trees, which switch among several well-defined behaviors in response to changing scenarios. The auctioning system is run at a fixed rate, allowing for newly added tasks to be incorporated into the auctioning system, which makes the solution reactive and allows for rerouting of some agents (subject to the states of the behavior trees). We demonstrate that the proposed architecture is especially well-suited for multistage tasks, where high costs are incurred when rerouting agents who have completed one or more stages of their current tasks. The proposed framework is experimentally validated in multiple scenarios in a lab environment. A video of a demonstration can be found at: https://***/ZdEkoOOlB2g.

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：