检索结果-内蒙古大学图书馆

From Missing Pieces to Masterpieces: Image Completion with Context-Adaptive Diffusion

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Shamsolmoali, Pourya Zareapoor, Masoumeh Zhou, Huiyu Felsberg, Michael Tao, Dacheng Li, Xuelong Department of Computer Science University of York University of UK SCEE East China Normal University China Department of Electrical Engineering Shanghai Jiao Tong University China School of Computing and Mathematical Sciences University of Leicester United Kingdom Computer Vision Laboratory Linkoping University Sweden College of Computing and Data Science Nanyang Technological University Singapore of China Telecom China

Image completion is a challenging task, particularly when ensuring that generated content seamlessly integrates with existing parts of an image. While recent diffusion models have shown promise, they often struggle with maintaining coherence between known and unknown (missing) regions. This issue arises from the lack of explicit spatial and semantic alignment during the diffusion process, resulting in content that does not smoothly integrate with the original image. Additionally, diffusion models typically rely on global learned distributions rather than localized features, leading to inconsistencies between the generated and existing image parts. In this work, we propose ConFill, a novel framework that introduces a Context-Adaptive Discrepancy (CAD) model to ensure that intermediate distributions of known and unknown regions are closely aligned throughout the diffusion process. By incorporating CAD, our model progressively reduces discrepancies between generated and original images at each diffusion step, leading to contextually aligned completion. Moreover, ConFill uses a new Dynamic Sampling mechanism that adaptively increases the sampling rate in regions with high reconstruction complexity. This approach enables precise adjustments, enhancing detail and integration in restored areas. Extensive experiments demonstrate that ConFill outperforms current methods, setting a new benchmark in image completion. © 2025, CC BY-SA.

关键词： Photointerpretation

Joint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jiang, Xiaoheng Yan, Feng Lu, Yang Wang, Ke Guo, Shuai Zhang, Tianzhu Pang, Yanwei Niu, Jianwei Xu, Mingliang The School of Computer and Artificial Intelligence Zhengzhou University Engineering Research Center of Intelligent Swarm Systems Ministry of Education National Supercomputing Center in Zhengzhou Zhengzhou450001 China The School of Computer and Artificial Intelligence Zhengzhou University Zhengzhou450001 China The School of Information Science and Technology University of Science and Technology of China Hefei230026 China The School of Electrical Automation and Information Engineering Tianjin University Tianjin300072 China The State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China Hangzhou Innovation Institute of Beihang University Hangzhou310051 China

Surface defect inspection plays an important role in the process of industrial manufacture and production. Though Convolutional Neural Network (CNN) based defect inspection methods have made huge leaps, they still confront a lot of challenges such as defect scale variation, complex background, low contrast, and so on. To address these issues, we propose a joint attention-guided feature fusion network (JAFFNet) for saliency detection of surface defects based on the encoder-decoder network. JAFFNet mainly incorporates a joint attention-guided feature fusion (JAFF) module into decoding stages to adaptively fuse low-level and high-level features. The JAFF module learns to emphasize defect features and suppress background noise during feature fusion, which is beneficial for detecting low-contrast defects. In addition, JAFFNet introduces a dense receptive field (DRF) module following the encoder to capture features with rich context information, which helps detect defects of different scales. The JAFF module mainly utilizes a learned joint channel-spatial attention map provided by high-level semantic features to guide feature fusion. The attention map makes the model pay more attention to defect features. The DRF module utilizes a sequence of multi-receptive-field (MRF) units with each taking as inputs all the preceding MRF feature maps and the original input. The obtained DRF features capture rich context information with a large range of receptive fields. Extensive experiments conducted on SD-saliency-900, Magnetic tile, and DAGM 2007 indicate that our method achieves promising performance in comparison with other state-of-the-art methods. Meanwhile, our method reaches a real-time defect detection speed of 66 FPS. Copyright © 2024, The Authors. All rights reserved.

关键词： Feature extraction

Robust Consensus of Perturbed Euler-Lagrange Agents with Unknown Disturbances

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2024年第6期58卷 190-195页

作者： Alexey Bobtsov Romeo Ortega Jose Guadalupe Romero Emmanuel Nuño Department of Control Systems and Robotics ITMO University Kronverkskiy av. 49 197101 Saint Petersburg Russia Departament of Electrical and Electronic Engineering ITAM Progreso Tizapán 1 04100 Ciudad de México México Department of Computer Science CUCEI University of Guadalajara 44430 Guadalajara Mexico

In this paper we consider the problems of leaderless consensus for networks of fully actuated Euler-Lagrange agents perturbed by unknown additive disturbances. The network is an undirected weighted graph with time delays. The proposed controller has a PD structure that incorporates, in a certainty-equivalent way, the estimate of the unknown disturbance. The design of the disturbance estimator proceeds along the following steps. First, the derivation of a regression equation, that turns out to be nonlinearly parameterized, but with an injective mapping. Second, we propose to use a recently introduced least-squares plus dynamic regressor extension algorithm that allows us to estimate the unknown frequencies imposing extremely weak excitation assumptions. In this way, we derive a sufficient condition on the proportional and derivative gains of the controller to ensure that the systems globally and asymptotically converge to a consensus position.

关键词： Robust adaptive control Multi-agent systems Disturbance rejection Robust time delay systems

Multi-Stage Knowledge Integration of Vision-Language Models for Continual Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Hongsheng Ji, Zhong Liu, Jingren Pang, Yanwei Han, Jungong The School of Electrical and Information Engineering The Tianjin Key Laboratory of Brain-Inspired Intelligence Technology Tianjin University Tianjin300072 China The Shanghai Artificial Intelligence Laboratory Shanghai200232 China The Department of Computer Science The University of Sheffield Sheffield United Kingdom

Vision Language Models (VLMs), pre-trained on large-scale image-text datasets, enable zero-shot predictions for unseen data but may underperform on specific unseen tasks. Continual learning (CL) can help VLMs effectively adapt to new data distributions without joint training, but faces challenges of catastrophic forgetting and generalization forgetting. Although significant progress has been achieved by distillation-based methods, they exhibit two severe limitations. One is the popularly adopted single-teacher paradigm fails to impart comprehensive knowledge, The other is the existing methods inadequately leverage the multimodal information in the original training dataset, instead they rely on additional data for distillation, which increases computational and storage overhead. To mitigate both limitations, by drawing on Knowledge Integration Theory (KIT), we propose a Multi-Stage Knowledge Integration network (MulKI) to emulate the human learning process in distillation methods. MulKI achieves this through four stages, including Eliciting Ideas, Adding New Ideas, Distinguishing Ideas, and Making Connections. During the four stages, we first leverage prototypes to align across modalities, eliciting cross-modal knowledge, then adding new knowledge by constructing fine-grained intra- and inter-modality relationships with prototypes. After that, knowledge from two teacher models is adaptively distinguished and re-weighted. Finally, we connect between models from intra- and inter-task, integrating preceding and new knowledge. Our method demonstrates significant improvements in maintaining zero-shot capabilities while supporting continual learning across diverse downstream tasks, showcasing its potential in adapting VLMs to evolving data distributions. © 2024, CC BY.

关键词： Zero-shot learning

Convolutional Learning on Multigraphs

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Butler, Landon Parada-Mayorga, Alejandro Ribeiro, Alejandro The Department of Electrical Engineering and Computer Science University of California BerkeleyCA94709 United States The Department of Electrical and Systems Engineering University of Pennsylvania PhiladelphiaPA19104 United States

Graph convolutional learning has led to many exciting discoveries in diverse areas. However, in some applications, traditional graphs are insufficient to capture the structure and intricacies of the data. In such scenarios, multigraphs arise naturally as discrete structures in which complex dynamics can be embedded. In this paper, we develop convolutional information processing on multigraphs and introduce convolutional multigraph neural networks (MGNNs). To capture the complex dynamics of information diffusion within and across each of the multigraph’s classes of edges, we formalize a convolutional signal processing model, defining the notions of signals, filtering, and frequency representations on multigraphs. Leveraging this model, we develop a multigraph learning architecture, including a generalization of selection sampling to reduce computational complexity. The introduced architecture is applied towards optimal wireless resource allocation and a hate speech localization task, offering improved performance over traditional graph neural networks. © 2022, CC BY.

关键词： Convolution

Generative AI for Data Augmentation in Wireless Networks: Analysis, Applications, and Case Study

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wen, Jinbo Kang, Jiawen Niyato, Dusit Zhang, Yang Wang, Jiacheng Sikdar, Biplab Zhang, Ping College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China School of Automation Guangdong University of Technology China College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering College of Design and Engineering National University of Singapore Singapore State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China

Data augmentation is a powerful technique to mitigate data scarcity. However, owing to fundamental differences in wireless data structures, traditional data augmentation techniques may not be suitable for wireless data. Fortunately, Generative Artificial Intelligence (GenAI) can be an effective alternative to wireless data augmentation due to its excellent data generation capability. This article systemically explores the potential and effectiveness of GenAI-driven data augmentation in wireless networks. We first briefly review data augmentation techniques, discuss their limitations in wireless networks, and introduce generative data augmentation, including reviewing GenAI models and their applications in data augmentation. We then explore the application prospects of GenAI-driven data augmentation in wireless networks from the physical, network, and application layers, which provides a GenAI-driven data augmentation architecture for each application. Subsequently, we propose a general generative diffusion model-based data augmentation framework for Wi-Fi gesture recognition, which uses transformer-based diffusion models to generate high-quality channel state information data. Furthermore, we develop residual neural network models for Wi-Fi gesture recognition to evaluate the role of augmented data and conduct a case study based on a real dataset. Simulation results demonstrate the effectiveness of the proposed framework. Finally, we discuss research directions for generative data augmentation. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

Bayesian Meta Regression 1

学校读者我要写书评

暂无评论

16th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2020 in conjunction with the 13th International Conference on Frontiers of Information Technology, Applications and Tools, FITAT 2020

作者： Munkhdalai, Lkhagvadorj Pham, Van Huy Ryu, Keun Ho Database/Bioinformatics Laboratory School of Electrical and Computer Engineering Chungbuk National University Cheongju28644 Korea Republic of Faculty of Information Technology Ton Duc Thang University Ho Chi Minh City700000 Viet Nam Department of Computer Science College of Electrical and Computer Engineering Chungbuk National University Cheongju28644 Korea Republic of

ISBN: (数字)9789813367579

ISBN: (纸本)9789813367562

This work extends Bayesian regression as an adaptive that augmented by deep neural networks (the probabilistic encoder) to obtain the posterior probability distributions of the regression coefficients. We use variational inference to obtain the conditional distribution over the regression coefficients, which are the latent space given the observed data. Therefore, our model can recognize local conditional probability distribution of the regression coefficients for each observation as well as it can measure the uncertainty of the model locally. Experimental results on the benchmark datasets prove that our Bayesian Meta Regression (BMR) significantly outperforms baseline regression techniques and precisely measures the model uncertainty for each observation. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.

关键词： Probability distributions

Do We Run Large-scale Multi-Robot systems on the Edge? More Evidence for Two-Phase Performance in System Size Scaling

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Kuckling, Jonas Luckey, Robin Avrutin, Viktor Vardy, Andrew Reina, Andreagiovanni Hamann, Heiko The Department of Computer and Information Science University of Konstanz Konstanz78457 Germany The Centre for the Advanced Study of Collective Behaviour University of Konstanz Germany The Institute of Computer Engineering University of Lübeck Lübeck Germany The Institute for Systems Theory and Automatic Control University of Stuttgart Stuttgart Germany The Department of Computer Science The Department of Electrical and Computer Engineering Memorial University of Newfoundland St. John's Canada IRIDIA Université Libre de Bruxelles Brussels Belgium

With increasing numbers of mobile robots arriving in real-world applications, more robots coexist in the same space, interact, and possibly collaborate. Methods to provide such systems with system size scalability are known, for example, from swarm robotics. Example strategies are self-organizing behavior, a strict decentralized approach, and limiting the robot-robot communication. Despite applying such strategies, any multi-robot system breaks above a certain critical system size (i.e., number of robots) as too many robots share a resource (e.g., space, communication channel). We provide additional evidence based on simulations, that at these critical system sizes, the system performance separates into two phases: nearly optimal and minimal performance. We speculate that in real-world applications that are configured for optimal system size, the supposedly high-performing system may actually live on borrowed time as it is on a transient to breakdown. We provide two modeling options (based on queueing theory and a population model) that may help to support this reasoning. © 2023, CC BY.

关键词： Multipurpose robots

Noise Robustness of Data Re-Uploading Quantum Classifiers

学校读者我要写书评

暂无评论

Noise Robustness of Data Re-Uploading Quantum Classifiers

Quantum Computing and engineering (QCE), IEEE International Conference on

作者： Daniel E. Molina Kathleen Hamilton Ming Hsieh Electrical and Computer Engineering Department University Of Southern California Los Angeles CA USA Computational Science and Engineering Division Oak Ridge National Laboratory Oak Ridge TN USA

We investigate the robustness of quantum neural networks (QNNs) trained as binary classifiers against bit flip and depolarizing noise, two channels that emulate a subset of noise present in noisy intermediate-scale quantum (NISQ) devices. The presence of decoherence and other forms of noise poses significant challenges to classification accuracy. To assess the impact of noise on label extraction, and to evaluate the performance and robustness of our models, we trained three different label assignment methods in the presence of various levels of noise. Each QNN model uses 2 qubits to implement a parameterized model and either the qubits of the QNN, or a set of ancilla qubits are used for label extraction. These three different methods project the state of either a single qubit or multiple qubits onto the computational basis. This generates a probability distribution that is post-processed to assign a class label and a class probability. Each QNN is trained using noisy numerical simulation on a dataset with 2 features. Depolarizing noise channels are interleaved with the parameterized gates of the QNN and we include bit flip error prior to extracting the labels. The noise levels for the depolarizing noise and bit flip error are swept over large ranges of values. We deploy QNNs trained at various noise levels on IBM quantum computers- this introduces an additional source of noise due to SWAP gate noise. We observe that the QNNs trained with simulated noise can have good performance on NISQ hardware.

关键词：