检索结果-内蒙古大学图书馆

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Lingyi Hong Shilin Yan Renrui Zhang Wanyun Li Xinyu Zhou Pinxue Guo Kaixun Jiang Yiting Chen Jinglun Li Zhaoyu Chen Wenqiang Zhang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China CUHK MMLab Shanghai Artificial Intelligence Laboratory Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai China Engineering Research Center of AI & Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Visual object tracking aims to localize the target object of each frame based on its initial appearance in the first frame. Depending on the input modility, tracking tasks can be divided into RGB tracking and RGB+X (e.g. RGB+N, and RGB+D) tracking. Despite the different input modalities, the core aspect of tracking is the temporal matching. Based on this common ground, we present a general framework to unify various tracking tasks, termed as One Tracker. One- Tracker first performs a large-scale pre-training on a RGB tracker called Foundation Tracker. This pretraining phase equips the Foundation Tracker with a stable ability to estimate the location of the target object. Then we regard other modality information as prompt and build Prompt Tracker upon Foundation Tracker. Through freezing the Foundation Tracker and only adjusting some additional trainable parameters, Prompt Tracker inhibits the strong localization ability from Foundation Tracker and achieves parameter- efficient finetuning on downstream RGB+X tracking tasks. To evaluate the effectiveness of our general framework OneTracker, which is consisted of Foundation Tracker and Prompt Tracker, we conduct extensive experiments on 6 popular tracking tasks across 11 benchmarks and our One- Tracker outperforms other models and achieves state-of-the-art performance.

关键词： Location awareness Visualization computer vision Adaptation models Target tracking Benchmark testing Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Whole Body Balance Control for Bipedal Robots Based on Virtual Model Control 16th

Whole Body Balance Control for Bipedal Robots Based on Virtu...

引用

16th International Conference on Intelligent robotics and Applications, ICIRA 2023

作者： Gao, Chengzhi Xie, Ye Zhu, Yiting Liang, Dingkun Kong, Lingyu Wang, Xin Xie, Anhuan Gu, Jianjun Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Zhejiang Laboratory Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Zhejiang University Zhejiang Hangzhou311100 China Department of Electrical and Computer Engineering Dalhousie University HalifaxNSB3H 4R2 Canada

ISBN: (纸本)9789819964918

Due to the lack of accurate stability theoretical basis, the dynamic balance control of bipedal robots is a challenging topic. In this paper, we propose a controller that combines virtual model control and whole body control. In our framework, the whole system is simplified into a spring inverted pendulum model. Decoupling control is carried out for this model, and a virtual model is introduced to establish the relationship between the joint torque of the support leg and the state of the center of mass. The distribution of reaction force is obtained through decoupling closed-loop control. The whole body control part first uses multi task whole body control to determine the kinematics information, and then solves the joint feedforward torque through dynamical model, thus obtaining the desired force position hybrid control. Proposed algorithm improved the problem of incomplete decoupling between states and increased the stiffness of the support leg. The newly designed control framework was tested in Simscape dynamic simulation and verified the anti-interference ability and fast walking ability of Whole Body Balance Control for bipedal robots based on Virtual Model Control. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Inverted pendulum

来源：评论

学校读者我要写书评

暂无评论

Decisioning-Based Approach for Optimising Control engineering Tools Using Digital Twin Capabilities and Other Cyber-Physical Metaverse Manufacturing System Components

引用

IEEE Transactions on Consumer Electronics 2024年第1期70卷 3212-3221页

作者： Mourad, Nahia Alsattar, Hassan A. Qahtan, Sarah Zaidan, Aws Alaa Deveci, Muhammet Sangaiah, Arun Kumar Pedrycz, Witold British University in Dubai Faculty of Engineering and IT Dubai United Arab Emirates University of Mashreq College of Administrative Science Department of Business Administration Baghdad10067 Iraq Middle Technical University College of Health and Medical Technology-Baghdad Department of Computer Center Baghdad10047 Iraq SP Jain School of Global Management SydneyNSW2141 Australia National Defence University Turkish Naval Academy Department of Industrial Engineering Istanbul34040 Turkey Lebanese American University Department of Electrical and Computer Engineering Byblos Lebanon International Graduate Institute of AI National Yunlin University of Science and Technology Yunlin64002 Taiwan University of Alberta Faculty of Engineering Department of Electrical and Computer Engineering EdmontonABT6G 1H9 Canada Systems Research Institute Polish Academy of Sciences Warsaw00-901 Poland Istinye University Faculty of Engineering and Natural Sciences Department of Computer Engineering Istanbul34010 Turkey

The optimisation of control engineering tools based on digital twin capabilities and other cyber-physical metaverse manufacturing system (CPMMS) components are crucial for the successful performance. This study proposes a model for optimising control engineering tools using digital twin capabilities and other CPMMS components to solve the open issues. The main contributions and novelty aspects of the methodological process are outlined as follows: Formulated and developed is a decision matrix based on a utility procedure for 10 control engineering tools with digital twin capabilities and other three CPMMS components (Programmable-Logic-Controller and Human-Machine-Interface, Internet of Things connectivity and cybersecurity features). This matrix accounts for the uncertainty associated with tool assessment and transformation evaluation issue;formulated and develop an integrating fuzzy weighted with zero-inconsistency-interval-valued spherical fuzzy rough sets (IvSFRS-FWZIC) and combined compromise solution (CoCoSo) methods. The IvSFRS-FWZIC method is utilised to assign importance degrees to the digital twin capabilities and other CPMMS components. The applicability and robustness of the proposed approach are validated and evaluated through conducting sensitivity, correlation, and comparative analyses. The proposed approach can assist managers in analysing and selecting the most suitable tool for developing CPMMS. © 1975-2011 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning

arXiv

引用

arXiv 2024年

作者： Hong, Lingyi Yan, Shilin Zhang, Renrui Li, Wanyun Zhou, Xinyu Guo, Pinxue Jiang, Kaixun Chen, Yiting Li, Jinglun Chen, Zhaoyu Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China The Chinese University of Hong Kong Hong Kong Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai China Engineering Research Center of AI & Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai China

Visual object tracking aims to localize the target object of each frame based on its initial appearance in the first frame. Depending on the input modility, tracking tasks can be divided into RGB tracking and RGB+X (e.g. RGB+N, and RGB+D) tracking. Despite the different input modalities, the core aspect of tracking is the temporal matching. Based on this common ground, we present a general framework to unify various tracking tasks, termed as OneTracker. OneTracker first performs a large-scale pre-training on a RGB tracker called Foundation Tracker. This pretraining phase equips the Foundation Tracker with a stable ability to estimate the location of the target object. Then we regard other modality information as prompt and build Prompt Tracker upon Foundation Tracker. Through freezing the Foundation Tracker and only adjusting some additional trainable parameters, Prompt Tracker inhibits the strong localization ability from Foundation Tracker and achieves parameter-efficient finetuning on downstream RGB+X tracking tasks. To evaluate the effectiveness of our general framework OneTracker, which is consisted of Foundation Tracker and Prompt Tracker, we conduct extensive experiments on 6 popular tracking tasks across 11 benchmarks and our OneTracker outperforms other models and achieves state-ofthe-art performance. Copyright © 2024, The Authors. All rights reserved.

关键词： Foundations

来源：评论

学校读者我要写书评

暂无评论

Blockchain Technology and Artificial Intelligence's Future Applications in Cyber Security

Blockchain Technology and Artificial Intelligence's Future A...

引用

Advancement in Electronics & Communication engineering (AECE), International Conference on

作者： Ramiz Salama Sinem Al-Turjman Chadi Altrjman Fadi Al-Turjman Richa Gupta Satya Prakash Yadav Satvik Vats Department of Computer Engineering AI and Robotics Institute Research Center for AI and IoT Near East University Nicosia TURKIYE Artificial Intelligence Engineering Dept AI and Robotics Institute Near East University Nicosia Turkey Department of Chemical Engineering Waterloo University ON Canada Department of Electronics & Communication Engineering Raj Kumar Goel Institute of technology Department of Computer Science and Engineering G.L. Bajaj Institute of Technology and Management(GLBITM) Greater Noida India Department of Computer Science and Engineering Graphic Era Hill University Dehradun India

Artificial intelligence and blockchain are quickly integrating in daily life and business applications. When numerous information systems must access and analyze data in real-time in centralized systems and applications, such as healthcare, a bottleneck develops. This issue would be resolved by blockchain's decentralized database architecture, safe data storage, data exchange, and authentication. Additionally, ai can be present at the very top of the blockchain and produce insights from the shared data that is created and applied to forecasts. Blockchain is a cutting-edge Cybersecurity technology that creates chains by mutual consent between nodes that chronologically link new blocks to those previously stored in nodes. A number of industries, including banking, insurance, cybersecurity, forecasting, healthcare, and cryptocurrencies, among others, are growing more quickly as a result of technology convergence. The likelihood of these systems being hacked increases as more digital technologies are used and these businesses provide services. Blockchain technology and artificial intelligence can work together to create a strong defense against these dangers and security problems. We'll examine how blockchain and ai are merged in cybersecurity in this chapter. We'll go into more detail about how they help secure cyber-physical systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Neural P3M: A Long-Range Interaction Modeling Enhancer for Geometric GNNs 38

Neural P3M: A Long-Range Interaction Modeling Enhancer for G...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wang, Yusong Cheng, Chaoran Li, Shaoning Ren, Yuxuan Shao, Bin Liu, Ge Heng, Pheng-Ann Zheng, Nanning 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 University of Illinois Urbana-Champaign United States Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong University of Science and Technology of China China Microsoft Research AI4Science United States

Geometric graph neural networks (GNNs) have emerged as powerful tools for modeling molecular geometry. However, they encounter limitations in effectively capturing long-range interactions in large molecular systems due to the localization assumption of GNN. To address this challenge, we introduce Neural P3M, a versatile enhancer of geometric GNNs to expand the scope of their capabilities by incorporating mesh points alongside atoms and reimaging traditional mathematical operations in a trainable manner. Neural P3M exhibits flexibility across a wide range of molecular systems and demonstrates remarkable accuracy in predicting energies and forces, outperforming on benchmarks such as the MD22 dataset. It also achieves an average improvement of 22% on the OE62 dataset while integrating with various architectures. Codes are available at https://***/OnlyLoveKFC/Neural_P3M. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LCGen: mining in low-certainty generation for view-consistent text-to-3D 24

LCGen: mining in low-certainty generation for view-consisten...

引用

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

作者： Zeng Tao Tong Yang Junxiong Lin Xinji Mai Haoran Wang Beining Wang Enyu Zhou Yan Wang Wenqiang Zhang Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China School of Computer Science Fudan University Shanghai China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China and Engineering Research Center of AI & Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai China

ISBN: (纸本)9798331314385

The Janus Problem is a common issue in SDS-based text-to-3D methods. Due to view encoding approach and 2D diffusion prior guidance, the 3D representation model tends to learn content with higher certainty from each perspective, leading to view inconsistency. In this work, we first model and analyze the problem, visualizing the specific causes of the Janus Problem, which are associated with discrete view encoding and shared priors in 2D lifting. Based on this, we further propose the LCGen method, which guides text-to-3D to obtain different priors with different certainty from various viewpoints, aiding in view-consistent generation. Experiments have proven that our LCGen method can be directly applied to different SDS-based text-to-3D methods, alleviating the Janus Problem without introducing additional information, increasing excessive training burden, or compromising the generation effect. Project page is https://***/zeng-tao/LCGen.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Scaling Diffusion Policy in Transformer to 1 Billion Parameters for Robotic Manipulation

arXiv

引用

arXiv 2024年

作者： Zhu, Minjie Zhu, Yichen Li, Jinming Wen, Junjie Xu, Zhiyuan Liu, Ning Cheng, Ran Shen, Chaomin Peng, Yaxin Feng, Feifei Tang, Jian School of Computer Science East China Normal University China Midea Group AI Research Center China Beijing Innovation Center of Humanoid Robotics China Department of Computer Science Shanghai University China

Diffusion Policy is a powerful technique tool for learning end-to-end visuomotor robot control. It is expected that Diffusion Policy possesses scalability, a key attribute for deep neural networks, typically suggesting that increasing model size would lead to enhanced performance. However, our observations indicate that Diffusion Policy in transformer architecture (DP-T) struggles to scale effectively;even minor additions of layers can deteriorate training outcomes. To address this issue, we introduce Scalable Diffusion Transformer Policy for visuomotor learning. Our proposed method, namely ScaleDP, introduces two modules that improve the training dynamic of Diffusion Policy and allow the network to better handle multimodal action distribution. First, we identify that DPT suffers from large gradient issues, making the optimization of Diffusion Policy unstable. To resolve this issue, we factorize the feature embedding of observation into multiple affine layers, and integrate it into the transformer blocks. Additionally, our utilize non-causal attention which allows the policy network to "see" future actions during prediction, helping to reduce compounding errors. We demonstrate that our proposed method successfully scales the Diffusion Policy from 10 million to 1 billion parameters. This new model, named ScaleDP, can effectively scale up the model size with improved performance and generalization. We benchmark ScaleDP across 50 different tasks from MetaWorld and find that our largest ScaleDP outperforms DP-T with an average improvement of 21.6%. Across 7 real-world robot tasks, our ScaleDP demonstrates an average improvement of 36.25% over DP-T on four single-arm tasks and 75% on three bimanual tasks. We believe our work paves the way for scaling up models for visuomotor learning. The project page is available at https://***/. © 2024, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Safety-Aware Trajectory Tracking using High-Order Control Barrier Functions

Safety-Aware Trajectory Tracking using High-Order Control Ba...

引用

IEEE Conference on Decision and Control

作者： Panagiotis Rousseas Dimitra Panagou Kostas Kyriakopoulos Control Systems Laboratory School of Mechanical Engineering National Technical University of Athens Greece Department of Robotics and the Department of Aerospace Engineering University of Michigan MI USA Center of AI & Robotics (CAIR) New York University Abu Dhabi

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

A novel method on safety-aware trajectory tracking is presented in this work. The proposed scheme employs Control Barrier Functions (CBFs) along with a latent-space representation over the barrier and its derivatives. While such representations have also been employed in the literature, the innovation of the proposed method lies in formulating a novel tracking problem on the latent space, enabling for the shape of linear class $-\mathcal{K}$ functions that are included in the safety condition induced by the control barrier function to be chosen constructively. Additionally, the aforementioned safety condition is modified based on the latent-space tracking through the addition of an extra term. This enables maintaining safety guarantees and is demonstrated to provide decreased tracking errors compared to the nominal control barrier function scheme. Numerical simulations validate the efficacy of the proposed scheme, as well as its ability for online implementation.

关键词： Technological innovation Trajectory tracking Shape Aerospace electronics Numerical simulation Safety

来源：评论

学校读者我要写书评

暂无评论

Attention in Attention for PET-CT Modality Consensus Lung Tumor Segmentation

Attention in Attention for PET-CT Modality Consensus Lung Tu...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yuzhou Zhao Xinyu Zhou Haijing Guo Qianyu Guo Yan Zuo Shaoli Song Shuyong Gao Wenqiang Zhang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Department of Nuclear Medicine Fudan University Shanghai Cancer Center China Shanghai Engineering Research Center of AI Robotics Academy for Engineering and Technology Fudan University

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Combination of multi-modal PET-CT imaging for lung tumor segmentation is significant for clinical treatment. Existing methods have not fully considered the impact of noise in PET-CT on the multi-modal interaction. To address this, we propose a novel Attention in Attention Network (aiANet). aiANet can mutually learn multi-modal characteristics for segmentation through its cross-learning modules. Within the cross-learning module, we introduce two nested-attention blocks, namely Attention in Self-Attention (aiSA) and Attention in Cross-Attention (aiCA), for multi-scale feature enhancement and multi-modal feature interaction. Importantly, since traditional attention weights calculated solely or unilaterally based on PET or CT can be vulnerable to the inevitable noisy information, we embed a novel Attention in Attention (aiA) module into aiSA and aiCA. The aiA module can seek cross-modal consensus for attention weights to alleviate their noise. Experimental results on clinical PET-CT data of lung cancer demonstrate the superiority of our method.

关键词： Image segmentation Computed tomography Aggregates Noise Lung cancer Imaging Noise measurement Tumors

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：