检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yu, Xuyi Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China

Due to the high complexity and technical requirements of industrial production processes, surface defects will inevitably appear, which seriously affects the quality of products. Although existing lightweight detection networks are highly efficient, they are susceptible to false or missed detection of non-salient defects due to the lack of semantic information. In contrast, the diffusion model can generate higher-order semantic representations in the denoising process. Therefore, the aim of this paper is to incorporate the higher-order modelling capability of the diffusion model into the detection model, so as to better assist in the classification and localization of difficult targets. First, the denoising diffusion probabilistic model (DDPM) is pretrained to extract the features of denoising process to construct as a feature repository. In particular, to avoid the potential bottleneck of memory caused by the dataloader loading high-dimensional features, a residual convolutional variational autoencoder (ResVAE) is designed to further compress the feature repository. The image is fed into both image backbone and feature repository for feature extraction and querying respectively. The queried latent features are reconstructed and filtered to obtain high-dimensional DDPM features. A dynamic cross-fusion method is proposed to fully refine the contextual features of DDPM to optimize the detection model. Finally, we employ knowledge distillation to migrate the higher-order modelling capabilities back into the lightweight baseline model without additional efficiency cost. Experiment results demonstrate that our method achieves competitive results on several industrial datasets. Copyright © 2024, The Authors. All rights reserved.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

Application of 3D Technology in Film and Television Animation System

Application of 3D Technology in Film and Television Animatio...

引用

作者： Zhang, Xiaoye Qiu, Sisi Altrjman, Chadi School of Comics Jilin Animation Institute Changchun130000 China Artificial Intelligence Engineering Department AI and Robotics Institute Near East University Mersin10 Turkey

In order to understand the application of 3D technology in film and television animation system, this paper will carry out relevant research. The study first introduces the basic concept and classification of 3D animation technology, then analyzes the key points of 3D technology application in each stage of film and television animation design, and discusses the application performance of 3D technology. Through research, 3D technology plays an important role in modern film and television animation system, and giving full play to the role of this technology can effectively improve the quality of animation and bring visual feast to people. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

EATNet: Efficient Axial Transformer Network for End-to-end Autonomous Driving

EATNet: Efficient Axial Transformer Network for End-to-end A...

引用

International Conference on Intelligent Transportation

作者： Weihuang Chen Fanjie Kong Liming Chen Shen'ao Wang Zhiping Wang Hongbin Sun Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China

ISBN: (数字)9798331505929

ISBN: (纸本)9798331505936

In recent years, end-to-end autonomous driving has garnered significant attention from researchers and has witnessed rapid advancements. However, existing methods en-counter challenges such as high computational demands, slow training and inference speeds, which hinder their real-world deployment. To tackle this issue, we introduce the Efficient Axial Transformer Network (EATNet), a lightweight multi-modal autonomous driving framework based on cross-axial Transformers. By effectively integrating lidar and multi-view RGB features, this model utilizes an enhanced lightweight cross-axial Transformer to minimize model size and computational requirements. Extensive experiments demonstrate that EATNet, with only a quarter of the parameters of comparable multi-modal models, achieves competitive or even superior performance on the closed-loop CARLA simulator compared to other baselines.

关键词： Training Human computer interaction Laser radar Computational modeling Transformers Feature extraction Natural language processing Computational efficiency Autonomous vehicles Intelligent transportation systems

来源：评论

学校读者我要写书评

暂无评论

Human-Like Dexterous Manipulation for the Anthropomorphic Hand-Arm Robotic System via Teleoperation 16th

Human-Like Dexterous Manipulation for the Anthropomorphic ...

引用

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

作者： Huang, Yayu Wang, Zhenghan Shen, Xiaofei Liu, Qian Wang, Peng Institute of Automation Chinese Academy of Sciences Beijing100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science and Innovation Chinese Academy of Sciences 999077 Hong Kong

ISBN: (纸本)9789819964918

Teleoperation has the potential to enable robots to replace humans in high-risk scenarios and catastrophic events, performing manipulation tasks efficiently and securely under human guidance. However, achieving human-like dexterous manipulation remains challenging, particularly for anthropomorphic hand-arm robotic systems with high degrees of freedom. Accurately capturing the operator’s motion and providing real-time intuitive feedback to enhance the sense of telepresence place substantial demands on human-robot perception and interaction. Moreover, the inherent physical and functional differences between anthropomorphic hand-arm robots and humans pose challenges in ensuring the accuracy and reliability of dexterous teleoperation. To overcome these challenges, we present an integrated approach involving an anthropomorphic hand-arm robot system, a wearable motion capture and force feedback system, and a set of motion mapping and force mapping methods. We conducted experiments on both simulation and real-world platforms to evaluate the usability and effectiveness of our proposed approach, with the results demonstrating significant advancements in achieving human-like dexterity via teleoperation. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

HyenaPixel: Global Image Context with Convolutions 27

HyenaPixel: Global Image Context with Convolutions

引用

27th European Conference on artificial intelligence, ECAI 2024

作者： Spravil, Julian Houben, Sebastian Behnke, Sven Fraunhofer IAIS Germany University of Applied Sciences Bonn-Rhein-Sieg Germany University of Bonn Computer Science Institute VI Center for Robotics Germany Lamarr Institute for Machine Learning and Artificial Intelligence Germany

ISBN: (纸本)9781643685489

In computer vision, a larger effective receptive field (ERF) is associated with better performance. While attention natively supports global context, its quadratic complexity limits its applicability to tasks that benefit from high-resolution input. In this work, we extend Hyena, a convolution-based attention replacement, from causal sequences to bidirectional data and two-dimensional image space. We scale Hyena's convolution kernels beyond the feature map size, up to 191×191, to maximize ERF while maintaining sub-quadratic complexity in the number of pixels. We integrate our two-dimensional Hyena, HyenaPixel, and bidirectional Hyena into the MetaFormer framework. For image categorization, HyenaPixel and bidirectional Hyena achieve a competitive ImageNet-1k top-1 accuracy of 84.9% and 85.2%, respectively, with no additional training data, while outperforming other convolutional and large-kernel networks. Combining HyenaPixel with attention further improves accuracy. We attribute the success of bidirectional Hyena to learning the data-dependent geometric arrangement of pixels without a fixed neighborhood definition. Experimental results on downstream tasks suggest that HyenaPixel with large filters and a fixed neighborhood leads to better localization performance. © 2024 The Authors.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Alignment of Traffic Elements for Domain Adaptive Semantic Segmentation 27

Unsupervised Alignment of Traffic Elements for Domain Adapti...

引用

27th IEEE International Conference on Intelligent Transportation Systems, ITSC 2024

作者： Gao, Yuan Li, Yaochen Zhang, Tengwen Qiu, Chao Liu, Yuehu School of Software Engineering Xi'an Jiaotong University Shaanxi Province Xi'an710049 China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Shaanxi Province Xi'an710049 China

ISBN: (纸本)9798331505929

Unsupervised domain adaptation has been used to reduce the domain shift, which would improve the performance of semantic segmentation on unlabeled real-world data. How-ever, existing methods do not cater to the specific characteristics of traffic scene elements, leading to suboptimal alignment outcomes. In this paper, we propose a novel domain adaptation method for semantic segmentation of road scenes via unsu-pervised alignment of traffic elements. Firstly, a self-training framework is developed, which distinguishes the differential features between dynamic and static traffic elements, providing more accurate alignment training. Then, we designed a dynamic and static traffic elements alignment module to achieve cross-domain feature matching between the source and the target domain images. The cosine similarity maximization is applied to the alignment of dynamic traffic elements, while the prototype learning is utilized for the static traffic elements. Furthermore, the element alignment loss functions for dynamic and static traffic elements are designed to optimize the alignment modules. The experimental results demonstrate that the proposed method is superior to the existing methods on G TAS→Cityscapes task, which is applicable to semantic segmentation of road scenes. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Tensegrity Hybrid Mobile Robot with Mechanical Impact Resistance 50

A Tensegrity Hybrid Mobile Robot with Mechanical Impact Resi...

引用

50th Annual Conference of the IEEE Industrial Electronics Society, IECON 2024

作者： Filimonov, Daniil Zhakataev, Altay Varol, Huseyin Atakan Robotics Engineering SEDS Nazarbayev University Astana Kazakhstan Mechanical Engineering SEDS Nazarbayev University Astana Kazakhstan Institute of Smart Systems and Artificial Intelligence Nazarbayev University Astana Kazakhstan

ISBN: (纸本)9781665464543

In this work, we present a hybrid mobile robot with legged and wheeled locomotion modes and tensegrity components. Our work uniquely integrates classical robot structures with tensegrities to create a lightweight robot capable of resisting fall impacts and moving efficiently. After describing the mechanical and embedded system implementation of the robot in detail, we present a set of experiments to demonstrate the efficacy of our robot. Specifically, we dropped the robot from different heights and conducted cost of transport measurements for wheeled and legged modes at different speeds. The findings demonstrate that the use of tensegrity structures could be beneficial in terms of impact resistance and in different challenging applications, while double-type locomotion system could serve for more efficient locomotion and power consumption. © 2024 IEEE.

关键词： Microrobots

来源：评论

学校读者我要写书评

暂无评论

Improving Sharpness-Aware Minimization by Lookahead 41

Improving Sharpness-Aware Minimization by Lookahead

引用

41st International Conference on Machine Learning, ICML 2024

作者： Yu, Runsheng Zhang, Youzhi Kwok, James T. Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation CAS Hong Kong

Sharpness-Aware Minimization (SAM), which performs gradient descent on adversarially perturbed weights, can improve generalization by identifying flatter minima. However, recent studies have shown that SAM may suffer from convergence instability and oscillate around saddle points, resulting in slow convergence and inferior performance. To address this problem, we propose the use of a lookahead mechanism to gather more information about the landscape by looking further ahead, and thus find a better trajectory to converge. By examining the nature of SAM, we simplify the extrapolation procedure, resulting in a more efficient algorithm. Theoretical results show that the proposed method converges to a stationary point and is less prone to saddle points. Experiments on standard benchmark datasets also verify that the proposed method outperforms the SOTAs, and converge more effectively to flat minima. Copyright 2024 by the author(s)

关键词：

来源：评论

学校读者我要写书评

暂无评论

引用

33rd ACM Web Conference, WWW 2024

作者： Ma, Changyi Yu, Runsheng Zhang, Youzhi Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science and Innovation CAS Hong Kong Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong

ISBN: (纸本)9798400701719

The similarity matrix is at the core of similarity search problems. However, incomplete observations are ubiquitous in real scenarios leading to a less accurate similarity matrix. To alleviate this problem, in this paper, based on the key insight that the similarity matrix enjoys both the symmetric and positive semi-definiteness (PSD) properties, we propose a novel similarity matrix calibration method, which is scalable, effective, and sound. Specifically, we establish the PSD property as a constraint for the similarity matrix calibration problem and propose a novel similarity matrix calibration method to estimate the similarity matrix, which approximates the unknown complete ground-truth similarity matrix. To enable a fast optimization process, we further develop a general approximated algorithm that bypasses the computation of singular values. Theoretical analysis ensures stable calibration performance and convergence speed. Extensive experiments of similarity matrix calibration on real-world datasets demonstrate that our proposed method outperforms baseline methods in terms of both accuracy and speed. © 2024 ACM.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Self-Consistency Training for Density-Functional-Theory Hamiltonian Prediction 41

Self-Consistency Training for Density-Functional-Theory Hami...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Zhang, He Liu, Chang Wang, Zun Wei, Xinran Liu, Siyuan Zheng, Nanning Shao, Bin Liu, Tie-Yan 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 Microsoft Research AI for Science United States

Predicting the mean-field Hamiltonian matrix in density functional theory is a fundamental formulation to leverage machine learning for solving molecular science ***, its applicability is limited by insufficient labeled data for *** this work, we highlight that Hamiltonian prediction possesses a self-consistency principle, based on which we propose self-consistency training, an exact training method that does not require labeled *** distinguishes the task from predicting other molecular properties by the following benefits: (1) it enables the model to be trained on a large amount of unlabeled data, hence addresses the data scarcity challenge and enhances generalization;(2) it is more efficient than running DFT to generate labels for supervised training, since it amortizes DFT calculation over a set of *** empirically demonstrate the better generalization in data-scarce and out-of-distribution scenarios, and the better efficiency over DFT *** benefits push forward the applicability of Hamiltonian prediction to an ever-larger scale. Copyright 2024 by the author(s)

关键词： Density functional theory

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：