检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Tian, Qingyao Chen, Zhen Liao, Huai Huang, Xinyan Li, Lujie Ourselin, Sebastien Liu, Hongbin Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Centre of AI and Robotics Hong Kong Institute of Science& Innovation Chinese Academy of Sciences Hong Kong The First Affiliated Hospital of Sun Yat-sen University Guangdong Guangzhou China School of Engineering and Imaging Sciences King’s College London United Kingdom Institute of Automation Chinese Academy of Sciences Centre of AI and Robotics Hong Kong Institute of Science& Innovation Chinese Academy of Sciences China

Single-image depth estimation is essential for endoscopy tasks such as localization, reconstruction, and augmented reality. Most existing methods in surgical scenes focus on in-domain depth estimation, limiting their real-world applicability. This constraint stems from the scarcity and inferior labeling quality of medical data for training. In this work, we present EndoOmni, the first foundation model for zero-shot cross-domain depth estimation for endoscopy. To harness the potential of diverse training data, we refine the advanced self-learning paradigm that employs a teacher model to generate pseudo-labels, guiding a student model trained on large-scale labeled and unlabeled data. To address training disturbance caused by inherent noise in depth labels, we propose a robust training framework that leverages both depth labels and estimated confidence from the teacher model to jointly guide the student model training. Moreover, we propose a weighted scale-and-shift invariant loss to adaptively adjust learning weights based on label confidence, thus imposing learning bias towards cleaner label pixels while reducing the influence of highly noisy pixels. Experiments on zero-shot relative depth estimation show that our EndoOmni improves state-of-the-art methods in medical imaging for 33% and existing foundation models for 34% in terms of absolute relative error on specific datasets. Furthermore, our model provides strong initialization for fine-tuning metric depth estimation, maintaining superior performance in both in-domain and out-of-domain scenarios. The source code is publicly available at https://github. com/TianCuteQY/EndoOmni. Copyright © 2024, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Interactive Navigation in Environments with Traversable Obstacles Using Large Language and Vision-Language Models

arXiv

引用

arXiv 2023年

作者： Zhang, Zhen Lin, Anran Wong, Chun Wai Chu, Xiangyu Dou, Qi Samuel Au, K.W. Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Hong Kong Multiscale Medical Robotics Centre Hong Kong Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Computer Aided Medical Procedures Technical University of Munich Germany

This paper proposes an interactive navigation framework by using large language and vision-language models, allowing robots to navigate in environments with traversable obstacles. We utilize the large language model (GPT-3.5) and the open-set Vision-language Model (Grounding DINO) to create an action-aware costmap to perform effective path planning without fine-tuning. With the large models, we can achieve an end-to-end system from textual instructions like "Can you pass through the curtains to deliver medicines to me?", to bounding boxes (e.g., curtains) with action-aware attributes. They can be used to segment LiDAR point clouds into two parts: traversable and untraversable parts, and then an action-aware costmap is constructed for generating a feasible path. The pre-trained large models have great generalization ability and do not require additional annotated data for training, allowing fast deployment in the interactive navigation tasks. We choose to use multiple traversable objects such as curtains and grasses for verification by instructing the robot to traverse them. Besides, traversing curtains in a medical scenario was tested. All experimental results demonstrated the proposed framework’s effectiveness and adaptability to diverse environments. Copyright © 2023, The Authors. All rights reserved.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Gesture Control Method for Oral Blood Mixed Liquid Removal Robot∗ 5

Gesture Control Method for Oral Blood Mixed Liquid Removal R...

引用

5th World Robot Conference Symposium on Advanced robotics and automation, WRC SARA 2023

作者： Ma, Xudong Kuang, Ziao Li, Han Chen, Anqi Li, Changsheng Liu, Wenyong Wang, Junchen Hu, Yida David Gao, Li Su, Bai-Quan Beijing University of Posts and Tele-communications Medical Robotics Laboratory School of Automation Beijing100876 China Beijing University of Posts and Tele-communications School of Computer Beijing100876 China Beijing Institute of Technology School of Mechatronical Engineering Beijing100081 China Beihang University Key Laboratory for Biomechanics and Mechanobiology of Chinese Education Ministry Beijing Advanced Innovation Centre for Biomedical Engineering School of Biological Science and Medical Engineering Beijing100083 China Beihang University School of Mechanical Engineering and Automation Beijing100083 China Brigham and Women's Hospital Harvard Medical School BostonMA02115 United States Peking University School and Hospital of Stomatology National Clinical Research Center for Oral Diseases National Stomatological Center Department of Periodontology Beijing China

ISBN: (纸本)9798350307320

All of oral surgeries are accompanied by bleeding. Manual removal of blood is carried out by a nurse and dozens of repetitive blood suction operations are implemented by the nurse during a single oral surgery. In our previous work, a robot for removal of blood-water mixed liquid is designed. However, the robot requires an initial order to autonomous removal of blood mixed liquid. In this paper, a hand gesture control method for typical operations of the robot is developed. Gesture dataset are developed at first and a deep learning neural network is trained for detection of hand gesture. The gesture control procedures for four typical operations are provided in detail. In a group of experiments, it's demonstrated that an oral surgeon controls the robot's motion by presenting prescribed hand gesture. The proposed method could be an alternative manner for blood mixed liquid removal. © 2023 IEEE.

关键词： Blood

来源：评论

学校读者我要写书评

暂无评论

Happy: a debiased learning framework for continual generalized category discovery 24

Happy: a debiased learning framework for continual generaliz...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Shijie Ma Fei Zhu Zhun Zhong Wenzhuo Liu Xu-Yao Zhang Cheng-Lin Liu MAIS Institute of Automation Chinese Academy of Sciences China and School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics HKISI-CAS China School of Computer Science and Information Engineering Hefei University of Technology China and School of Computer Science University of Nottingham NG8 1BB Nottingham UK

ISBN: (纸本)9798331314385

Constantly discovering novel concepts is crucial in evolving environments. This paper explores the underexplored task of Continual Generalized Category Discovery (C-GCD), which aims to incrementally discover new classes from unlabeled data while maintaining the ability to recognize previously learned classes. Although several settings are proposed to study the C-GCD task, they have limitations that do not reflect real-world scenarios. We thus study a more practical C-GCD setting, which includes more new classes to be discovered over a longer period, without storing samples of past classes. In C-GCD, the model is initially trained on labeled data of known classes, followed by multiple incremental stages where the model is fed with unlabeled data containing both old and new classes. The core challenge involves two conflicting objectives: discover new classes and prevent forgetting old ones. We delve into the conflicts and identify that models are susceptible to prediction bias and hardness bias. To address these issues, we introduce a debiased learning framework, namely Happy, characterized by Hardness-aware prototype sampling and soft entropy regularization. For the prediction bias, we first introduce clustering-guided initialization to provide robust features. In addition, we propose soft entropy regularization to assign appropriate probabilities to new classes, which can significantly enhance the clustering performance of new classes. For the harness bias, we present the hardness-aware prototype sampling, which can effectively reduce the forgetting issue for previously seen classes, especially for difficult classes. Experimental results demonstrate our method proficiently manages the conflicts of C-GCD and achieves remarkable performance across various datasets, e.g., 7.5% overall gains on ImageNet-100. Our code is publicly available at https://***/mashijie1028/Happy-CGCD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ASI-Seg: Audio-Driven Surgical Instrument Segmentation with Surgeon Intention Understanding

ASI-Seg: Audio-Driven Surgical Instrument Segmentation with ...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Zhen Chen Zongming Zhang Wenwu Guo Xingjian Luo Long Bai Jinlin Wu Hongliang Ren Hongbin Liu Centre for Artificial Intelligence and Robotics (CAIR) Hong Kong Institute of Science & Innovation Chinese Academy of Sciences China Department of Electronic Engineering Chinese University of Hong Kong China State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Surgical instrument segmentation is crucial in surgical scene understanding, thereby facilitating surgical safety. Existing algorithms directly detected all instruments of predefined categories in the input image, lacking the capability to segment specific instruments according to the surgeon’s intention. During different stages of surgery, surgeons exhibit varying preferences and focus toward different surgical instruments. Therefore, an instrument segmentation algorithm that adheres to the surgeon’s intention can minimize distractions from irrelevant instruments and assist surgeons to a great extent. The recent Segment Anything Model (SAM) reveals the capability to segment objects following prompts, but the manual annotations for prompts are impractical during the surgery. To address these limitations in operating rooms, we propose an audio-driven surgical instrument segmentation framework, named ASI-Seg, to accurately segment the required surgical instruments by parsing the audio commands of surgeons. Specifically, we propose an intention-oriented multimodal fusion to interpret the segmentation intention from audio commands and retrieve relevant instrument details to facilitate segmentation. Moreover, to guide our ASI-Seg segment of the required surgical instruments, we devise a contrastive learning prompt encoder to effectively distinguish the required instruments from the irrelevant ones. Therefore, our ASI-Seg promotes the workflow in the operating rooms, thereby providing targeted support and reducing the cognitive load on surgeons. Extensive experiments are performed to validate the ASI-Seg framework, which reveals remarkable advantages over classical state-of-the-art and medical SAMs in both semantic segmentation and intention-oriented segmentation. The source code is available at https://***/Zonmgin-Zhang/ASI-Seg.

关键词： Instruments Semantic segmentation Source coding Surgery Contrastive learning Manuals Cognitive load Safety Intelligent robots Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Mutual Impacts of Different Dead-times and Control Sample Times on SiC Inverter Loss in Experimental Motor Drive at High Carrier Frequencies

Mutual Impacts of Different Dead-times and Control Sample Ti...

引用

International Conference on Communications and Electronics, ICCE

作者： Nguyen Gia Minh Thao Van-Long Pham Kutsukake Asuka Keisuke Fujisaki Ton Duc Do Department of Mechanical Electrical and Electronic Engineering Interdisciplinary Faculty of Science and Engineering Shimane University Matsue Japan Department of Automation School of Electrical and Electronic Engineering Hanoi University of Science and Technology Hanoi Vietnam Next Generation Tatara Co-Creation Centre Shimane University Matsue Japan Electromagnetic Energy System Laboratory Toyota Technological Institute Nagoya Japan Department of Robotics and Mechatronics School of Engineering and Digital Sciences Nazarbayev University Astana Kazakhstan

ISBN: (数字)9798350379792

ISBN: (纸本)9798350379808

At high pulse-width modulation (PWM) carrier frequencies up to 200 kHz using silicon carbide (SiC) devices, the individual and mutual impacts of the sample time and dead-time on the SiC inverter loss in a motor drive system may become significantly. Thus, these issues under load condition are experimentally investigated and analyzed in the present study, where the dead-time is from 250 to 1000 ns and the control sample time is from 100 to 1000 μs. Furthermore, the measured results with the SiC metal–oxide-semiconductor field-effect transistor (MOSFET) inverter are briefly compared to that with the silicon (Si) insulated-gate bipolar transistor (IGBT) inverter for reference. Moreover, the total harmonics distortion (THD) of the measured voltage and current of the considered interior permanent magnet synchronous motor (IPMSM) is provided to explain the results. Finally, insights with theoretical analysis are shown as complementary explanations of the experimental findings. The outcome of this research can be treated as a reference in choosing the suitable value of the control sample time for IPMSM drive systems using SiC-MOSFET inverters.

关键词： Semiconductor device measurement Time-frequency analysis Motor drives Voltage measurement Magnetic field measurement Silicon carbide Current measurement

来源：评论

学校读者我要写书评

暂无评论

A Fuzzy Logic Controller for a Photovoltaic System Relying on a Closed-Loop DC-DC Converter

A Fuzzy Logic Controller for a Photovoltaic System Relying o...

引用

International Conference on science Technology engineering and Management (ICONSTEM)

作者： B. Chandrashekar Jayarama Pradeep Muthukumaran. M M. Sivasubramanian K. Lakshmi Khandan Joshuva Arockia Dhanraj Department of Computer Science Engineering St Martin's Engineering College Hyderabad Telangana India Department of Electrical and Electronics Engineering St. Joseph's College of Engineering Chennai Tamil Nadu India Department of Electrical and Electronics Engineering Karpagam college of Engineering Coimbatore Tamil Nadu Department of Electrical and Electronics Engineering Srivenkateswara College of Engineering and Technology Thiruvallur Tamil Nadu India Department of Electrical and Electronics Engineering Madanapalle Institute of Technology and Science Madanapalle Andhra Pradesh India Department of Mechatronics Engineering Centre for Automation and Robotics (ANRO) Hindustan Institute of Technology and Science Chennai Tamil Nadu India

ISBN: (数字)9798350365092

ISBN: (纸本)9798350365108

A DC-DC converter that can handle a lot of different power changes. The experts in this study discovered that adding a fuzzy logic circuit helped the system work better. It makes a big difference in how well and consistently the PV system works that the FLC controls the output power of the converter. Because the FLC works with Microsoft Excel, more people can use this powerful tool for less money and with less trouble. The FLC cuts down on “overshoot,” “settling time,” and steady-state error, as shown in simulations that test how well the controller works. This study may have made it easier for people to start making “fuzzy logic controllers” and make them better. This will have a big effect on how far we get in making energy systems that work better and are more stable. Further investigation could be conducted to determine whether or not these concepts are feasible and how they would enhance the FLC in different types. These DC-DC converters have closed loops in their circuitry.

关键词： Fuzzy logic Photovoltaic systems DC-DC power converters Control systems Energy efficiency Steady-state Spreadsheet programs

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE International Conference on Image Processing

作者： Chaowei Han Gaofeng Meng Chunlei Huo State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences School of Artificial Intelligence University of Chinese Academy of Sciences Centre for Artificial Intelligence and Robotics HK Institute of Science & Innovation Chinese Academy of Sciences School of Information Engineering Capital Normal University Beijing China

ISBN: (数字)9798350349399

ISBN: (纸本)9798350349405

Content-based video retrieval aims to retrieve near-duplicate entries from a database of a given query video. It plays an important role in combating video piracy. Robustness to video temporal dynamics is crucial for a representation model in video retrieval, as frames extracted from two copied videos are hardly temporally aligned in actual situations. However, current image retrieval datasets have difficulty in evaluating this robustness. To address this issue, we collect Similar Frame Dataset (SFD), which consists of 32,923 query-target pairs with 128,240 distraction images. The task of SFD is to retrieve the target frame from all items given a query frame. SFD is constructed by sampling frames from Kinetics-700 action classification dataset. An object detection model (Faster R-CNN) and a Multimodal Large Language Model (BLIP2) are used during sampling to select those valid frames. Besides, we propose Adjacent Frames Contrastive Learning (AFCL) framework. In AFCL, adjacent frames are sampled from unlabeled videos as positive pairs. An image representation model with robustness to changing frames can be trained under AFCL framework and achieve the state-of-the-art performance on SFD. The code will be released at https://***/Chuan-shanjia/Similar-Frame-Dataset.

关键词： Codes Databases Scalability Large language models Image retrieval Contrastive learning Object detection

来源：评论

学校读者我要写书评

暂无评论

Active Generalized Category Discovery

arXiv

引用

arXiv 2024年

作者： Ma, Shijie Zhu, Fei Zhong, Zhun Zhang, Xu-Yao Liu, Cheng-Lin MAIS Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics HKISI-CAS China School of Computer Science and Information Engineering Hefei University of Technology China School of Computer Science University of Nottingham NottinghamNG8 1BB United Kingdom

Generalized Category Discovery (GCD) is a pragmatic and challenging open-world task, which endeavors to cluster unlabeled samples from both novel and old classes, leveraging some labeled data of old classes. Given that knowledge learned from old classes is not fully transferable to new classes, and that novel categories are fully unlabeled, GCD inherently faces intractable problems, including imbalanced classification performance and inconsistent confidence between old and new classes, especially in the low-labeling regime. Hence, some annotations of new classes are deemed necessary. However, labeling new classes is extremely costly. To address this issue, we take the spirit of active learning and propose a new setting called Active Generalized Category Discovery (AGCD). The goal is to improve the performance of GCD by actively selecting a limited amount of valuable samples for labeling from the oracle. To solve this problem, we devise an adaptive sampling strategy, which jointly considers novelty, informativeness and diversity to adaptively select novel samples with proper uncertainty. However, owing to the varied orderings of label indices caused by the clustering of novel classes, the queried labels are not directly applicable to subsequent training. To overcome this issue, we further propose a stable label mapping algorithm that transforms ground truth labels to the label space of the classifier, thereby ensuring consistent training across different active selection stages. Our method achieves state-of-the-art performance on both generic and fine-grained datasets. Our code is available at https://github. com/mashijie1028/ActiveGCD Copyright © 2024, The Authors. All rights reserved.

关键词： Conformal mapping

来源：评论

学校读者我要写书评

暂无评论

Happy: A Debiased Learning Framework for Continual Generalized Category Discovery

arXiv

引用

arXiv 2024年

作者： Ma, Shijie Zhu, Fei Zhong, Zhun Liu, Wenzhuo Zhang, Xu-Yao Liu, Cheng-Lin MAIS Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics HKISI-CAS China School of Computer Science and Information Engineering Hefei University of Technology China School of Computer Science University of Nottingham NottinghamNG8 1BB United Kingdom

关键词： Federated learning

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：