检索结果-内蒙古大学图书馆

Modality Unified Attack for Omni-Modality Person Re-Identification

IEEE Transactions on information Forensics and Security 2025年

作者： Bian, Yuan Liu, Min Yi, Yunqi Wang, Xueping Ma, Yunfeng Wang, Yaonan Hunan University National Engineering Research Center of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha China Hunan Normal University Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing College of Information Science and Engineering Hunan Changsha China

Deep learning based person re-identification (re-id) models have been widely employed in surveillance systems. Recent studies have demonstrated that black-box single-modality and cross-modality re-id models are vulnerable to adversarial examples (AEs), leaving the robustness of multi-modality re-id models unexplored. Due to the lack of knowledge about the specific type of model deployed in the target black-box surveillance system, we aim to generate modality unified AEs for omni-modality (single-, cross- and multi-modality) re-id models. Specifically, we propose a novel Modality Unified Attack method to train modality-specific adversarial generators to generate AEs that effectively attack different omni-modality models. A multi-modality model is adopted as the surrogate model, wherein the features of each modality are perturbed by metric disruption loss before fusion. To collapse the common features of omnimodality models, Cross Modality Simulated Disruption approach is introduced to mimic the cross-modality feature embeddings by intentionally feeding images to non-corresponding modality-specific subnetworks of the surrogate model. Moreover, Multi Modality Collaborative Disruption strategy is devised to facilitate the attacker to comprehensively corrupt the informative content of person images by leveraging a multi modality feature collaborative metric disruption loss. Extensive experiments show that our MUA method can effectively attack the omni-modality re-id models, achieving 55.9%, 24.4%, 49.0% and 62.7% mean mAP Drop Rate, respectively. © 2005-2012 IEEE.

关键词： Modality unified attack Multi Modality Collaborative Disruption Omni-Modality re-id

来源：评论

学校读者我要写书评

暂无评论

PARE-Net: Position-Aware Rotation-Equivariant Networks for Robust Point Cloud Registration

arXiv

引用

arXiv 2024年

作者： Yao, Runzhao Du, Shaoyi Cui, Wenting Tang, Canhui Yang, Chengwu 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

Learning rotation-invariant distinctive features is a fundamental requirement for point cloud registration. Existing methods often use rotation-sensitive networks to extract features, while employing rotation augmentation to learn an approximate invariant mapping rudely. This makes networks fragile to rotations, overweight, and hinders the distinctiveness of features. To tackle these problems, we propose a novel position-aware rotation-equivariant network, for efficient, light-weighted, and robust registration. The network can provide a strong model inductive bias to learn rotation-equivariant/invariant features, thus addressing the aforementioned limitations. To further improve the distinctiveness of descriptors, we propose a position-aware convolution, which can better learn spatial information of local structures. Moreover, we also propose a feature-based hypothesis proposer. It leverages rotation-equivariant features that encode fine-grained structure orientations to generate reliable model hypotheses. Each correspondence can generate a hypothesis, thus it is more efficient than classic estimators that require multiple reliable correspondences. Accordingly, a contrastive rotation loss is presented to enhance the robustness of rotation-equivariant features against data degradation. Extensive experiments on indoor and outdoor datasets demonstrate that our method significantly outperforms the SOTA methods in terms of registration recall while being lightweight and keeping a fast speed. Moreover, experiments on rotated datasets demonstrate its robustness against rotation variations. Code is available at https://***/yaorz97/PARENet. Copyright © 2024, The Authors. All rights reserved.

关键词： HTTP

来源：评论

学校读者我要写书评

暂无评论

PMT: Progressive Mean Teacher via Exploring Temporal Consistency for Semi-Supervised Medical Image Segmentation

arXiv

引用

arXiv 2024年

作者： Gao, Ning Zhou, Sanping Wang, Le 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

Semi-supervised learning has emerged as a widely adopted technique in the field of medical image segmentation. The existing works either focuses on the construction of consistency constraints or the generation of pseudo labels to provide high-quality supervisory signals, whose main challenge mainly comes from how to keep the continuous improvement of model capabilities. In this paper, we propose a simple yet effective semi-supervised learning framework, termed Progressive Mean Teachers (PMT), for medical image segmentation, whose goal is to generate high-fidelity pseudo labels by learning robust and diverse features in the training process. Specifically, our PMT employs a standard mean teacher to penalize the consistency of the current state and utilizes two sets of MT architectures for co-training. The two sets of MT architectures are individually updated for prolonged periods to maintain stable model diversity established through performance gaps generated by iteration differences. Additionally, a difference-driven alignment regularizer is employed to expedite the alignment of lagging models with the representation capabilities of leading models. Furthermore, a simple yet effective pseudo-label filtering algorithm is employed for facile evaluation of models and selection of high-fidelity pseudo-labels outputted when models are operating at high performance for co-training purposes. Experimental results on two datasets with different modalities, i.e., CT and MRI, demonstrate that our method outperforms the state-of-the-art medical image segmentation approaches across various dimensions. The code is available at https://***/Axi404/PMT. © 2024, CC BY.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Learnable Frequency Filters and Low-Frequency Domain Randomization Based Domain Generalization for Medical Image Segmentation

Learnable Frequency Filters and Low-Frequency Domain Randomi...

引用

2023 China Automation Congress, CAC 2023

作者： Chen, Zikai Wu, Jiayi Li, Jincheng Zuo, Weiliang Li, Chenzdong Xin, Jingmin Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Xi'an China College of Clinical Medicine Medical College of Henan University of Science and Technology The First Affiliated Hospital Luoyang China

ISBN: (纸本)9798350303759

In recent years, the convolutional neural network represented by U-net has achieved great success in the field of medical image segmentation. However, due to different data acquisition protocols and differences in instrument parameters, the data in actual clinical applications will have a distribution shift problem with the training data of the model, also known as the domain shift problem. This can lead to a decrease in model performance, hindering the application of the model in real-world clinical settings. Traditional convolutional neural networks have local perception properties, and their inability to model long-range dependencies hinders the improvement of domain generalization performance. In addition, the traditional DG method of feature space domain randomization is not effective in medical image segmentation tasks due to the limited search space of feature styles and the inability to keep semantic information unchanged. To address the above issues, this paper proposes a medical image segmentation network named GFDR-U net for domain generalization. The network utilizes globally learnable filters to model long-range dependencies, and introduces a low-frequency domain randomization method to solve the problem of semantic invariance. Extensive experiments show that our method outperforms the state-of-the-art methods. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Improved Capon Estimator for High-Resolution DOA Estimation and Its Statistical Analysis

Improved Capon Estimator for High-Resolution DOA Estimation ...

引用

作者： Zuo, Weiliang Xin, Jingmin Liu, Changnong Zheng, Nanning Sano, Akira National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Xi'an710049 China Data Center China Construction Bank Beijing100033 China Keio University Department of System Design Engineering Yokohama223-8522 Japan

Despite some efforts and attempts have been made to improve the direction-of-arrival (DOA) estimation performance of the standard Capon beamformer (SCB) in array processing, rigorous statistical performance analyses of these modified Capon estimators are still lacking. This paper studies an improved Capon estimator (ICE) for estimating the DOAs of multiple uncorrelated narrowband signals, where the higher-order inverse (sample) array covariance matrix is used in the Capon-like cost function. By establishing the relationship between this nonparametric estimator and the parametric and classic subspace-based MUSIC (multiple signal classification), it is clarified that as long as the power order of the inverse covariance matrix is increased to reduce the influence of signal subspace components in the ICE, the estimation performance of the ICE becomes equivalent to that of the MUSIC regardless of the signal-to-noise ratio (SNR). Furthermore the statistical performance of the ICE is analyzed, and the large-sample mean-squared-error (MSE) expression of the estimated DOA is derived. Finally the effectiveness and the theoretical analysis of the ICE are substantiated through numerical examples, where the Cramer-Rao lower bound (CRB) is used to evaluate the validity of the derived asymptotic MSE expression. © 2014 Chinese Association of Automation.

关键词： Multiple signal classification

来源：评论

学校读者我要写书评

暂无评论

EventAid: Benchmarking Event-aided Image/Video Enhancement Algorithms with Real-captured Hybrid Dataset

arXiv

引用

arXiv 2023年

作者： Duan, Peiqi Li, Boyu Yang, Yixin Lou, Hanyue Teng, Minggui Ma, Yi Shi, Boxin National Key Laboratory for Multimedia Information Processing National Engineering Research Center of Visual Technology School of Computer Science Peking University China

—Event cameras are emerging imaging technology that offers advantages over conventional frame-based imaging sensors in dynamic range and sensing speed. Complementing the rich texture and color perception of traditional image frames, the hybrid camera system of event and frame-based cameras enables high-performance imaging. With the assistance of event cameras, high-quality image/video enhancement methods make it possible to break the limits of traditional frame-based cameras, especially exposure time, resolution, dynamic range, and frame rate limits. This paper focuses on five event-aided image and video enhancement tasks (i.e., event-based video reconstruction, event-aided high frame rate video reconstruction, image deblurring, image super-resolution, and high dynamic range image reconstruction), provides an analysis of the effects of different event properties, a real-captured and ground truth labeled benchmark dataset, a unified benchmarking of state-of-the-art methods, and an evaluation for two mainstream event simulators. In detail, this paper collects a real-captured evaluation dataset EVENTAID for five event-aided image/video enhancement tasks, by using "Event-RGB" multi-camera hybrid system, taking into account scene diversity and spatiotemporal synchronization. We further perform quantitative and visual comparisons for state-of-the-art algorithms, provide a controlled experiment to analyze the performance limit of event-aided image deblurring methods, and discuss open problems to inspire future research. Copyright © 2023, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Nearly Lossless Adaptive Bit Switching

arXiv

引用

arXiv 2025年

作者： Huang, Haiduo Liu, Zhenhua Xia, Tian Zhao, Wenzhe Ren, Pengju The National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi Xi’an China

Model quantization is widely applied for compressing and accelerating deep neural networks (DNNs). However, conventional Quantization-Aware Training (QAT) focuses on training DNNs with uniform bit-width. The bit-width settings vary across different hardware and transmission demands, which induces considerable training and storage costs. Hence, the scheme of one-shot joint training multiple precisions is proposed to address this issue. Previous works either store a larger FP32 model to switch between different precision models for higher accuracy or store a smaller INT8 model but compromise accuracy due to using shared quantization parameters. In this paper, we introduce the Double Rounding quantization method, which fully utilizes the quantized representation range to accomplish nearly lossless bit-switching while reducing storage by using the highest integer precision instead of full precision. Furthermore, we observe a competitive interference among different precisions during one-shot joint training, primarily due to inconsistent gradients of quantization scales during backward propagation. To tackle this problem, we propose an Adaptive Learning Rate Scaling (ALRS) technique that dynamically adapts learning rates for various precisions to optimize the training process. Additionally, we extend our Double Rounding to one-shot mixed precision training and develop a Hessian-Aware Stochastic Bit-switching (HASB) strategy. Experimental results on the ImageNet-1K classification demonstrate that our methods have enough advantages to state-of-the-art one-shot joint QAT in both multi-precision and mixed-precision. We also validate the feasibility of our method on detection and segmentation tasks, as well as on LLMs task. Our codes are available at https://***/haiduo/Double-Rounding. Copyright © 2025, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Generating Realistic VRU-AV Scenarios via Social Force-Based Gradient Optimization 27

Generating Realistic VRU-AV Scenarios via Social Force-Based...

引用

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

作者： Long, Tingting Zong, Mengru Zhang, Chi Liu, Yuehu Ma, Shuangxun Li, Li National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Shaanxi Xi'an China School of Electrical Engineering Xi' An Jiaotong University Shaanxi Xi'an China School of Future Transportation Changan University Xi'an710064 China BNRist Tsinghua University Department of Automation Beijing100084 China

ISBN: (纸本)9798331505929

Validating safe and reliable interaction with vulnerable road users (VRUs) is imperative for autonomous vehicles, especially in simulation testing procedures before on-road applications. However, existing VRU agents are either with predefined routes or with collision-oriented behaviors, which limits the fidelity and diversity of scenarios. Bridging the gap between the simulated world and reality requires realistic VRU behaviors. In this paper, we apply social force to capture natural human behavior, considering how individuals are influenced by their surroundings and adhere to real-world rules. To this end, we propose HybridSFM, an improved social force model designed for characterizing human-vehicle interacting behavior. Specifically, the steering and acceleration of the VRUs are parameterized and integrated with social norms to forecast trajectories. Based on this, we further implement a gradient optimization framework for generating VRU-AV safety-critical scenarios. Firstly, VRUs are positioned randomly around the ego-agent. Differentiable simulation updates states and unrolls trajectories while backpropagation optimizes the actions of VRUs. The modified scenario is further optimized in the subsequent time step until a collision happens. The experiments demonstrate that our approach can enhance realistic safety-critical scenarios at intersection scenarios, facilitating the testing of interaction capabilities for autonomous vehicles. © 2024 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Estimating Correlation Between Brain Consciousness and Depth of Anesthesia Based on EEG

Estimating Correlation Between Brain Consciousness and Depth...

引用

2021 China Automation Congress, CAC 2021

作者： Zhang, Wen Ma, Yongqiang Xin, Jingmin Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an710049 China National Engineering Laboratory for Visual Information Processing and Applications Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9781665426473

Electroencephalography (EEG) is an intuitive indicator forjudging the patient's level of consciousness and the depth of anesthesia (DOA) in clinical surgery. Evaluating the patient's DOA according to EEG can assist the anesthesiologist to more accurately determine the type and dose of anesthetics. However, due to the irregularity of EEG waveform, it is difficult to directly give the correlation between the patient's level of consciousness and DOA. From this point of view, in this article, the nonlinear dynamic methods and the frequency-domain methods are combined to analyze EEG, and characteristic parameters are used to analyze the changing trend of the patient's level of consciousness during general anesthesia. Then the Elman Neural Network (ElmanNN) and the Long and Short-Term Memory Neural Network (LSTM) are used to extract the temporal features of the characteristic sequence segments and identify the patient's DOA in the current period of time and predict the DOA in the next period of time. The results show that in the recognition and prediction of the DOA, when the aforementioned characteristic indicators are input as the combined parameter, the test accuracy of the ElmanNN is 88.46% and 82.69%, and the test accuracy of the LSTM is 90.38% and 74.04%. Compared with other feature combination methods, this model is more accurate in determining the patient's DOA. © 2021 IEEE

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Multi-AGV pathfinding for automatic warehouse applications

Multi-AGV pathfinding for automatic warehouse applications

引用

2021 China Automation Congress, CAC 2021

作者： Tao, Lesong Zhang, Songyi Chen, Shitao Zheng, Nanning Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an710049 China National Engineering Laboratory for Visual Information Processing and Applications Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9781665426473

Optimal coordination and scheduling of multiple automatic guided vehicles (AGVs) is one of the key issues of automatic warehouse systems. However, the existing algorithms perform unsatisfyingly on the large-scale number of agents and the map size due to the excessive time required for optimal scheduling. This paper proposes an alternate iterative conflict-based search (AICBS) algorithm. In this algorithm, path extension and conflict resolution are performed alternately, that is, checking whether there is a conflict at each step of the extension, and if there is a conflict modifying the routes of the two conflicting agents to avoid invalid plans caused by the path extension after the conflict, so as to save search time. In addition, the algorithm also reduces the time cost of conflict resolution by fixing the priority of agents on the vertexes and edges of the graph which is constructed from the map of the warehouse. The algorithm can be implemented on a general topology map, not only on a raster map, so that the time cost can be further reduced by compressing the map size. We compare the proposed algorithm with the conflict-based search algorithm in a raster map and apply it to a general topology map. Experimental results show that the algorithm can effectively reduce the time for task coordination and scheduling of multi-AGV. © 2021 IEEE

关键词： Multi agent systems

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：