检索结果-内蒙古大学图书馆

4th International Conference on robotics and computer Vision, ICRCV 2022

作者： Yu, Zhenghong Wang, Hao Yang, Hang Zhou, Huabing Guangdong Polytechnic of Science and Technology School of Robotics Zhuhai China Hubei Key Laboratory of Intelligent Robot Wuhan Institute of Technology College of Computer Science and Engineering Wuhan Institute of Technology Wuhan China

ISBN: (数字)9781665481700

ISBN: (纸本)9781665481700

As computer vision and human-computer interaction technologies are getting mature, vision-based auxiliary text reading has become the mainstream method to optimize learning and reading experience. To overcome the drawbacks of the common multi-step detection way and further improve the text reading experience, we propose a new human-centered fast text-locating method, named HTD. It transforms the relatively time-consuming serial procedure to a hybrid object detection problem, utilizing the hand-text hybrid object detection model to instantly locate the text of interest to readers. Considering the particularity of this task, we choose to perform the weighted NMS to infer final predicted bounding box. To verify the effectiveness of the proposed method, we tested the performance on the homemade hand-text hybrid object detection dataset HTDD. The experiment indicated that our proposed method is more efficiency and can produce satisfying results, which could meet the needs of human-computer interactive auxiliary reading in text reading scenarios. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Diffusion Trajectory-guided Policy for Long-horizon Robot Manipulation

arXiv

引用

arXiv 2025年

作者： Fan, Shichao Yang, Quantao Liu, Yajie Wu, Kun Che, Zhengping Liu, Qingjie Wan, Min School of Mechanical Engineering and Automation BeiHang University China School of Computer Science and Engineering BeiHang University China Beijing Innovation Center of Humanoid Robotics China KTH Royal Institute of Technology Sweden

Recently, Vision-Language-Action models (VLA) have advanced robot imitation learning, but high data collection costs and limited demonstrations hinder generalization and current imitation learning methods struggle in out-of-distribution scenarios, especially for long-horizon tasks. A key challenge is how to mitigate compounding errors in imitation learning, which lead to cascading failures over extended trajectories. To address these challenges, we propose the Diffusion Trajectory-guided Policy (DTP) framework, which generates 2D trajectories through a diffusion model to guide policy learning for long-horizon tasks. By leveraging task-relevant trajectories, DTP provides trajectory-level guidance to reduce error accumulation. Our two-stage approach first trains a generative vision-language model to create diffusion-based trajectories, then refines the imitation policy using them. Experiments on the CALVIN benchmark show that DTP outperforms state-of-the-art baselines by 25% in success rate, starting from scratch without external pretraining. Moreover, DTP significantly improves real-world robot performance. All source code and experiment videos will be released upon acceptance. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Task-Oriented Co-Design of Communication, Computing, and Control for Edge-Enabled Industrial Cyber-Physical Systems

arXiv

引用

arXiv 2025年

作者： Diao, Yufeng Zhang, Yichi De Martini, Daniele Zhao, Philip Guodong Li, Emma Liying School of Computing Science University of Glasgow United Kingdom Department of Computer Science University of Manchester United Kingdom Oxford Robotics Institute Department of Engineering Science University of Oxford United Kingdom

This paper proposes a task-oriented co-design framework that integrates communication, computing, and control to address the key challenges of bandwidth limitations, noise interference, and latency in mission-critical industrial Cyber-Physical Systems (CPS). To improve communication efficiency and robustness, we design a task-oriented Joint Source-Channel Coding (JSCC) using Information Bottleneck (IB) to enhance data transmission efficiency by prioritizing task-specific information. To mitigate the perceived End-to-End (E2E) delays, we develop a Delay-Aware Trajectory-Guided Control Prediction (DTCP) strategy that integrates trajectory planning with control prediction, predicting commands based on E2E delay. Moreover, the DTCP is co-designed with task-oriented JSCC, focusing on transmitting task-specific information for timely and reliable autonomous driving. Experimental results in the CARLA simulator demonstrate that, under an E2E delay of 1 second (20 time slots), the proposed framework achieves a driving score of 48.12, which is 31.59 points higher than using Better Portable Graphics (BPG) while reducing bandwidth usage by 99.19%. Copyright © 2025, The Authors. All rights reserved.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Integrated CWT-CNN for Epilepsy Detection Using Multiclass EEG Dataset

引用

computers, Materials & Continua 2021年第10期69卷 471-486页

作者： Sidra Naseem Kashif Javed Muhammad Jawad Khan Saddaf Rubab Muhammad Attique Khan Yunyoung Nam Robotics and Machine Intelligence Engineering School of Mechanical and Manufacturing EngineeringNational University of Sciences and Technology(NUST)Islamabad44000Pakistan National University of Sciences and Technology(NUST) Islamabad44000Pakistan Department of Computer Science HITEC University TaxilaTaxilaPakistan Department of Computer Science and Engineering Soonchunhyang UniversityAsanKorea

Electroencephalography is a common clinical procedure to record brain signals generated by human *** are useful in Brain controlled interfaces and other intelligent Neuroscience applications,but manual analysis of these brainwaves is complicated and time-consuming even for the experts of *** EEG analysis and classification techniques have been proposed to address this problem however,the conventional classification methods require identification and learning of specific EEG characteristics *** learning models can learn features from data without having in depth knowledge of data and prior feature *** of the great implementations of deep learning is Convolutional Neural Network(CNN)which has outperformed traditional neural networks in pattern recognition and image *** Wavelet Transform(CWT)is an efficient signal analysis technique that presents the magnitude of EEG signals as timerelated Frequency *** deep learning architectures suffer from poor performance when classifying EEG signals in the Time-frequency *** improve classification accuracy,we propose an integrated CWT and CNN technique which classifies five types of EEG signals *** compared the results of proposed integrated CWT and CNN method with existing deep learning models e.g.,GoogleNet,VGG16,***,the accuracy and loss of the proposed integrated CWT and CNN method have been cross validated using Kfold cross *** average accuracy and loss of Kfold cross-validation for proposed integrated CWT and CNN method are,76.12%and 56.02%*** model produces results on a publicly available dataset:Epilepsy dataset by UCI(Machine Learning Repository).

关键词： Deep learning electroencephalography epilepsy continuous wavelet transform

来源：评论

学校读者我要写书评

暂无评论

Towards combating frequency simplicity-biased learning for domain generalization 24

Towards combating frequency simplicity-biased learning for d...

引用

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

作者： Xilin He Jingyu Hu Qinliang Lin Cheng Luo Weicheng Xie Siyang Song Muhammad Haris Khan Linlin Shen Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University and Shenzhen Institute of Artificial Intelligence and Robotics for Society and Guangdong Provincial Key Laboratory of Intelligent Information Processing University of Exeter Mohamed bin Zayed University of Artificial Intelligence

ISBN: (纸本)9798331314385

Domain generalization methods aim to learn transferable knowledge from source domains that can generalize well to unseen target domains. Recent studies show that neural networks frequently suffer from a simplicity-biased learning behavior which leads to over-reliance on specific frequency sets, namely as frequency shortcuts, instead of semantic information, resulting in poor generalization performance. Despite previous data augmentation techniques successfully enhancing generalization performances, they intend to apply more frequency shortcuts, thereby causing hallucinations of generalization improvement. In this paper, we aim to prevent such learning behavior of applying frequency shortcuts from a data-driven perspective. Given the theoretical justification of models' biased learning behavior on different spatial frequency components, which is based on the dataset frequency properties, we argue that the learning behavior on various frequency components could be manipulated by changing the dataset statistical structure in the Fourier domain. Intuitively, as frequency shortcuts are hidden in the dominant and highly dependent frequencies of dataset structure, dynamically perturbating the over-reliance frequency components could prevent the application of frequency shortcuts. To this end, we propose two effective data augmentation modules designed to collaboratively and adaptively adjust the frequency characteristic of the dataset, aiming to dynamically influence the learning behavior of the model and ultimately serving as a strategy to mitigate shortcut learning. Code is available at https://***/C0notSilly/AdvFrequency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

VinT-6D: a large-scale object-in-hand dataset from vision, touch and proprioception 24

VinT-6D: a large-scale object-in-hand dataset from vision, t...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Zhaoliang Wan Yonggen Ling Senlin Yi Lu Qi Wangwei Lee Minglei Lu Sicheng Yang Xiao Teng Peng Lu Xu Yang Ming-Hsuan Yang Hui Cheng School of Computer Science and Engineering Sun Yat-sen University Guangzhou China and Robotics X Tencnet Shenzhen China Robotics X Tencnet Shenzhen China School of Computer Science and Engineering Sun Yat-sen University Guangzhou China The University of California Merced Merced Chinese Academic of Sciences Automation Institute Beijing China

This paper addresses the scarcity of large-scale datasets for accurate object-in-hand pose estimation, which is crucial for robotic in-hand manipulation within the "Perception-Planning-Control" paradigm. Specifically, we introduce VinT-6D, the first extensive multi-modal dataset integrating vision, touch, and proprioception, to enhance robotic manipulation. VinT-6D comprises 2 million VinT-Sim and 0.1 million VinT-Real splits, collected via simulations in MuJoCo and Blender and a custom-designed real-world platform. This dataset is tailored for robotic hands, offering models with whole-hand tactile perception and highquality, well-aligned data. To the best of our knowledge, the VinT-Real is the largest considering the collection difficulties in the real-world environment so that it can bridge the gap of simulation to real compared to the previous works. Built upon VinT-6D, we present a benchmark method that shows significant improvements in performance by fusing multi-modal information. The project is available at https://***/.

关键词：

来源：评论

学校读者我要写书评

暂无评论

TAGIC: Task-Guided Image Communication Framework for Seamless Teleoperation

TAGIC: Task-Guided Image Communication Framework for Seamles...

引用

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Yufeng Diao Yichi Zhang Philip G. Zhao Daniele De Martini School of Engineering University of Glasgow UK Department of Computer Science University of Manchester UK Oxford Robotics Institute (ORI) University of Oxford UK

ISBN: (数字)9798350384475

ISBN: (纸本)9798350384482

Image-based teleoperation offers significant flexibility and efficiency in several applications, such as teleoperated driving; still, it highly depends on reliable communication band-width and high Signal-to-Noise Ratio (SNR), which is hard to guarantee in uncontrolled environments. This poster tackles the challenge of reliable communication under limited bandwidth. We propose to leverage the context and task knowledge to guide the compression to favor task performance rather than image fidelity. In particular, we jointly designed source-channel coding with a task performer to present an end-to-end TAsk-Guided Image Communication (TAGIC) framework, which uses Soft Introspective Variational Autoencoder (S-IntroVAE) and prioritizes the task-critical image information with limited communication bandwidth in the low SNR region. We demonstrate the effectiveness of TAGIC in a teleoperated driving scenario through the CARLA simulation platform - a widely used simulator in the autonomous driving community. Given the equivalent value of bandwidth compression ratio, TAGIC achieves a 202.6% improvement in the driving score over existing methods at low SNR.

关键词： Resistance Knowledge engineering Image coding Image communication Conferences Bandwidth Reliability

来源：评论

学校读者我要写书评

暂无评论

Remote Sensing Scene Classification Based on Hybrid Convolutional Neural Network and Attention Module

Remote Sensing Scene Classification Based on Hybrid Convolut...

引用

Advancement in Data science, E-learning and Information Systems (ICADEIS), International Conference on

作者： Muhammad Yusuf Kardawi Alexander Agung Santoso Gunawan Profesional Engineer Program Faculty of Engineering Bina Nusantara University Jakarta Indonesia Computer Science Department Faculty of Computer Science University of Indonesia Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331513320

ISBN: (纸本)9798331513337

Remote sensing images present classification challenges due to the complexity of their structural and spatial patterns. This research explores a hybrid approach that combines convolutional neural network (CNN) and attention through feature fusion to improve scene classification accuracy in remote sensing images. The proposed architecture utilizes EfficientNet and VGGNet to extract depth features separately. The extracted features are then integrated with Dynamic Selfattention (DSA), which dynamically focuses the model on the most relevant information in the image. DSA allows the model to adaptively assign weights to different parts of the image, thus improving the discriminative ability of the model. Furthermore, a feature fusion technique is applied to combine information from different layers of the CNN and DSA modules. Experiments conducted on the UC Merced dataset showed accuracies of 0.9181 and 0.9167. These results show that the combination of CNN, DSA, and feature fusion is an effective and robust approach for remote sensing image classification.

关键词： Adaptation models Scene classification Accuracy Electronic learning Data science Feature extraction Convolutional neural networks Data mining Remote sensing Information systems

来源：评论

学校读者我要写书评

暂无评论

India’s Stake in Emerging Technology - A Review

India’s Stake in Emerging Technology - A Review

引用

International Conference on Computing for Sustainable Global Development (INDIACom)

作者： Sugandha Singh S China Ramu Ruchika Lalit Department of Computer Science and Engineering SGT University Gurugram India Department of Computer Science and Engineering Chaitanya Bharathi Institute of Technology Hyderabad India University School of Automation and Robotics GGSIPU East Campus New Delhi India

Emerging technologies of today are altering the way we think about the built environment. In the meantime, many technologies have undergone revolutionary changes, and India, an expanding nation, offers opportunities for development in several new areas. The goal of the study is to show how organizations need to adapt their perspective and create new strategies for emerging technologies such as artificial intelligence (AI), machine learning (ML), big data, quantum, and space exploration in light of the swift advancements in technology. This paper presents that although India has come across a lot of advancements in this era of emerging technology, certain challenges have been faced in aspects like adapting, feasibility, and approval which brought many changes across the country, which in turn is impacting the pace to transform the country towards Industry 4.0 and Industry 5.0. Therefore a study is presented to understand the impact of emerging technologies for countries coming up as Industry 4.0 and Industry 5.0. This paper emphasizes on for India’s initiatives for emerging technologies to come up as VIKSIT BHARAT @2047.

关键词：

来源：评论

学校读者我要写书评

暂无评论

APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： He, Weizhao Zhang, Yang Zhuo, Wei Shen, Linlin Yang, Jiaqi Deng, Songhe Sun, Liang Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China School of Computer Science University of Nottingham China

Few-shot semantic segmentation (FSS) endeavors to segment unseen classes with only a few labeled samples. Current FSS methods are commonly built on the assumption that their training and application scenarios share similar domains, and their performances degrade significantly while applied to a distinct domain. To this end, we propose to leverage the cutting-edge foundation model, the Segment Anything Model (SAM), for generalization enhancement. The SAM however performs unsatisfactorily on domains that are distinct from its training data, which primarily comprise natural scene images, and it does not support automatic segmentation of specific semantics due to its interactive prompting mechanism. In our work, we introduce APSeg, a novel auto-prompt network for cross-domain few-shot semantic segmentation (CD-FSS), which is designed to be auto-prompted for guiding cross-domain segmentation. Specifically, we propose a Dual Prototype Anchor Transformation (DPAT) module that fuses pseudo query prototypes extracted based on cycle-consistency with support prototypes, allowing features to be transformed into a more stable domain-agnostic space. Additionally, a Meta Prompt Generator (MPG) module is introduced to automatically generate prompt embeddings, eliminating the need for manual visual prompts. We build an efficient model which can be applied directly to target domains without fine-tuning. Extensive experiments on four cross-domain datasets show that our model outperforms the state-of-the-art CD-FSS method by 5.24% and 3.10% in average accuracy on 1-shot and 5-shot settings, respectively. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

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：