检索结果-内蒙古大学图书馆

A survey on federated learning:a perspective from multi-party computation

Frontiers of Computer Science 2024年第1期18卷 93-103页

作者： Fengxia LIU Zhiming ZHENG Yexuan SHI Yongxin TONG Yi ZHANG Institute of Artificial Intelligence and Key Laboratory of Mathematics Informatics Behavioral Semantics Beihang UniversityBeijing 100191China State Key Laboratory of Software Development Environment and Advanced Innovation Center for Future Blockchain and Privacy Computing Beihang UniversityBeijing 100191China Pengcheng Laboratory Shenzhen 518055China Zhongguancun Laboratory Beijing 100190China Institute for Mathematical Sciences and Engineering Research Center of Financial Computing and Digital Engineering Renmin University of ChinaBeijing 100872China

Federated learning is a promising learning paradigm that allows collaborative training of models across multiple data owners without sharing their raw *** enhance privacy in federated learning,multi-party computation can be leveraged for secure communication and computation during model *** survey provides a comprehensive review on how to integrate mainstream multi-party computation techniques into diverse federated learning setups for guaranteed privacy,as well as the corresponding optimization techniques to improve model accuracy and training *** also pinpoint future directions to deploy federated learning to a wider range of applications.

关键词： sfederated learning multi-party ycomputation privacy-preserving data mining distributed learning

来源：评论

学校读者我要写书评

暂无评论

A New Adversarial Perspective for LiDAR-Based 3D Object Detection 39

A New Adversarial Perspective for LiDAR-Based 3D Object Dete...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Zheng, Shijun Liu, Weiquan Guo, Yu Zang, Yu Shen, Siqi Wang, Cheng Fujian Key Laboratory of Sensing and Computing for Smart Cities School of Informatics Xiamen University China Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China School of Informatics Xiamen University China College of Computer Engineering Jimei University China

ISBN: (纸本)157735897X

Autonomous vehicles (AVs) rely on LiDAR sensors for environmental perception and decision-making in driving scenarios. However, ensuring the safety and reliability of AVs in complex environments remains a pressing challenge. To address this issue, we introduce a real-world dataset (ROLiD) comprising LiDAR-scanned point clouds of two random objects: water mist and smoke. In this paper, we introduce a novel adversarial perspective by proposing an attack framework that utilizes water mist and smoke to simulate environmental interference. Specifically, we propose a point cloud sequence generation method using a motion and content decomposition generative adversarial network named PCS-GAN to simulate the distribution of random objects. Furthermore, leveraging the simulated LiDAR scanning characteristics implemented with Range Image, we examine the effects of introducing random object perturbations at various positions on the target vehicle. Extensive experiments demonstrate that adversarial perturbations based on random objects effectively deceive vehicle detection and reduce the recognition rate of 3D object detection models. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Vehicle detection

来源：评论

学校读者我要写书评

暂无评论

Text to Point Cloud Localization with Multi-Level Negative Contrastive Learning 39

Text to Point Cloud Localization with Multi-Level Negative C...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Liu, Dunqiang Huang, Shujun Li, Wen Shen, Siqi Wang, Cheng Fujian Key Laboratory of Sensing and Computing for Smart Cities Xiamen University China Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China School of Informatics Xiamen University China

ISBN: (纸本)157735897X

Language-based localization is a crucial task in robotics and computer vision, enabling robots to understand spatial positions through language. Recent methods rely on contrastive learning to establish correspondences between global features of texts and point clouds. However, the inherent ambiguity of textual descriptions makes it difficult to convey geometric information accurately, forcing alignment of them in the feature space may compromise the expressiveness of the point clouds. Unlike previous methods, this paper proposes using language as a filter to distinguish dissimilar locations. To this end, we propose a robust framework of multi-level negative contrastive learning for language-based localization, fully leveraging the descriptive power of language for spatial localization. Our method learns multiple mismatched factors by minimizing the similarity of different locations at different levels, including global-level, instance-level and relation-level, respectively. Extensive experiments conducted on the KITTI360Pose benchmark demonstrate that our method outperforms better that the state-of-the-art methods. Specifically, we achieve a 56.3% improvement in Top-1 retrieval recall and a 45.9% improvement in 5m localization recall. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

HROF: a high-resolution remote sensing dataset for segmentation of offshore farms 5

HROF: a high-resolution remote sensing dataset for segmentat...

引用

5th International Conference on Geoscience and Remote Sensing Mapping, ICGRSM 2023

作者： Qiu, Rihuang Cheng, Ming Fujian Key Laboratory of Sensing and Computing for Smart Cities School of Informatics Xiamen University Fujian Xiamen361005 China

ISBN: (纸本)9781510672789

Deep learning (DL) methods have achieved promising results in high resolution (HR) remote sensing image (RSI) segmentation. While DL approaches rely heavily on large-scale training datasets, and a method trained on a dataset generally cannot be used to segment classes that the dataset does not contain. Existing RSI datasets mainly cover buildings, roads, vegetation, water, moving objects, etc. In this paper, we focus on a class of objects that have been rarely involved in previous datasets: offshore farms. The monitoring and management of offshore farms is important for the sustainable development of the aquaculture industry, and in this work, we provide an HR remote sensing dataset of offshore farms, named HROF. This dataset uses two types of sensors, i.e., multispectral and synthetic aperture radar (SAR). The multispectral data consists of 30 images with a resolution of about 1 m, including both raft and long line farms. The SAR data contains 41 images with a resolution of 3 m, in which the objects are breeding ponds. Objects in each image are manually annotated at the pixel-precise level. Furthermore, we provide baseline segmentation method and results on HROF and compare with mainstream methods. The dataset is available at https://***/FrontierQ/HROF. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Improving Occluded Human Pose Estimation Via Linked Joints 48

Improving Occluded Human Pose Estimation Via Linked Joints

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Ye, Suhang Hong, Zebo Zheng, Jiawen Zhang, Shengchuan Xiamen University Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics China

ISBN: (纸本)9781728163277

Keypoint heatmaps, which produce peak values of Gaussian distributions for individual human joints, are crucial components in 2D human pose estimation. However, existing merits using keypoint heatmaps are usually defeated in body occlusion, resulting in inaccurate joint predictions. We consider that the failure is mainly due to keypoint heatmaps' insufficiency for distinguishing the joints from two occluded bodies. Therefore, in this paper, we propose a method termed SkeletonMap (SMap), which introduces the prior knowledge of body structure to constrain relative connection of joints. As an extension of keypoint heatmaps, SMap can be efficiently plugged into existing 2D human pose estimation models with negligible increase in computational cost. Extensive experiments are conducted to show the effectiveness and generalization of SMap. Without bells and whistles, SMap brings a significant performance boost to the existing heatmap-based 2D human pose estimation models. On the MPII dataset, SMap improves SimpleBaseline (ResNet-152) from89.7 PCKh@0.5 to 90.4, and HRNet (W32) from 90.5 to 90.9. On the COCO dataset, SMap improves SimpleBaseline (ResNet-152) from 72.4 AP to 73.9. On the more challenging OCHuman dataset, SMap improves HRNet (W32) from 61.9 AP to 64.5, achieving 2.6 AP gains. We hope our simple and efficient approach will serve as a solid component for future research in 2D human pose estimation. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Visible-Infrared Person Re-identification via Modality Unification and Confidence Guidance 24

Semi-supervised Visible-Infrared Person Re-identification vi...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zheng, Xiying Zhang, Yukang Lu, Yang Wang, Hanzi Fujian Key Laboratory of Sensing and Computing for Smart City School of Informatics Xiamen University Xiamen China Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University Xiamen China

ISBN: (纸本)9798400706868

Semi-supervised visible-infrared person re-identification (SSVI-ReID) aims to match pedestrian images of the same identity from different modalities (visible and infrared) while only annotating visible images, which is highly related to multimedia and multi-modal processing. Existing works primarily focus on assigning accurate pseudo-labels to infrared images, but overlook the two key challenges: erroneous pseudo-labels and large modality discrepancy. To alleviate these issues, this paper proposes a novel Modality-Unified and Confidence-Guided (MUCG) semi-supervised learning method. Specifically, we first propose a Dynamic Intermediate Modality Generation (DIMG) module, which transfers knowledge from labeled visible images to unlabeled infrared images, enhancing the pseudo-label quality and bridging the modality discrepancy. Meanwhile, we propose a Weighted Identification Loss (WIL) that can reduce the model's dependence on erroneous labels by using confidence weighting. Moreover, an effective Modality Consistency Loss (MCL) is proposed to narrow the distribution of visible and infrared features, further narrowing the modality discrepancy and enabling the learning of modality-unified features. Extensive experiments show that the proposed MUCG has significant advantages in improving the performance of the SSVI-ReID task, surpassing the current state-of-the-art methods by a significant margin. © 2024 ACM.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

GLATrack: Global and Local Awareness for Open-Vocabulary Multiple Object Tracking 24

GLATrack: Global and Local Awareness for Open-Vocabulary Mul...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Li, Guangyao Jian, Yajun Yan, Yan Wang, Hanzi Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University Xiamen China Fujian Key Laboratory of Sensing and Computing for Smart City School of Informatics Xiamen University Xiamen China

ISBN: (纸本)9798400706868

Open-vocabulary multi-object tracking (MOT) aims to track arbitrary objects encountered in the real world beyond the training set. However, recent methods rely solely on instance-level detection and association of novel objects, which may not consider the valuable fine-grained semantic representations of the targets within key and reference frames. In this paper, we propose a Global and Local Awareness open-vocabulary MOT method (GLATrack), which learns to tackle the task of real-world MOT from both global and instance-level perspectives. Specifically, we introduce a region-aware feature enhancement module to refine global knowledge for complementing local target information, which enhances semantic representation and bridges the distribution gap between the image feature map and the pooled regional features. We propose a bidirectional semantic complementarity strategy to mitigate semantic misalignment arising from missing target information in key frames, which dynamically selects valuable information within reference frames to enrich object representation during the knowledge distillation process. Furthermore, we introduce an appearance richness measurement module to provide appropriate representations for targets with different appearances. The proposed method gains an improvement of 6.9% in TETA and 5.6% in mAP on the large-scale TAO benchmark. © 2024 ACM.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

X-Oscar: A Progressive Framework for High-quality Text-guided 3D Animatable Avatar Generation 41

X-Oscar: A Progressive Framework for High-quality Text-guide...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Ma, Yiwei Lin, Zhekai Ji, Jiayi Fan, Yijun Sun, Xiaoshuai Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China School of Informatics Xiamen University 361005 China

Recent advancements in automatic 3D avatar generation guided by text have made significant progress. However, existing methods have limitations such as oversaturation and low-quality output. To address these challenges, we propose X-Oscar, a progressive framework for generating high-quality animatable avatars from text prompts. It follows a sequential "Geometry→Texture→Animation" paradigm, simplifying optimization through step-by-step generation. To tackle oversaturation, we introduce Adaptive Variational Parameter (AVP), representing avatars as an adaptive distribution during training. Additionally, we present Avatar-aware Score Distillation Sampling (ASDS), a novel technique that incorporates avatar-aware noise into rendered images for improved generation quality during optimization. Extensive evaluations confirm the superiority of X-Oscar over existing text-to-3D and text-to-avatar approaches. Our project page is https://***/Projects/XOscar/. Copyright 2024 by the author(s)

关键词： 3D modeling

来源：评论

学校读者我要写书评

暂无评论

Joint Active and Passive Beamforming Design in Intelligent Reflecting Surface(IRS)-Assisted Covert Communications:A Multi-Agent DRL Approach

引用

China Communications 2024年第9期21卷 11-26页

作者： Gao Ang Ren Xiaoyu Deng Bin Sun Xinshun Zhang Jiankang School of Electronic Information Northwestern Polytechnical UniversityXi’an 710072China Key Laboratory of Near Ground Detection and Perception Technology Wuxi 214035China Department of Computing and Informatics Bournemouth UniversityUK

Intelligent Reflecting Surface(IRS),with the potential capability to reconstruct the electromagnetic propagation environment,evolves a new IRSassisted covert communications paradigm to eliminate the negligible detection of malicious eavesdroppers by coherently beaming the scattered signals and suppressing the signals ***,when multiple IRSs are involved,accurate channel estimation is still a challenge due to the extra hardware complexity and communication *** the crossinterference caused by massive reflecting paths,it is hard to obtain the close-formed solution for the optimization of covert *** this basis,the paper improves a heterogeneous multi-agent deep deterministic policy gradient(MADDPG)approach for the joint active and passive beamforming(Joint A&P BF)optimization without the channel estimation,where the base station(BS)and multiple IRSs are taken as different types of agents and learn to enhance the covert spectrum efficiency(CSE)*** to the‘centralized training and distributed execution’feature of MADDPG,each agent can execute the active or passive beamforming independently based on its partial observation without referring to *** results demonstrate that the proposed deep reinforcement learning(DRL)approach could not only obtain a preferable CSE of legitimate users and a low detection of probability(LPD)of warden,but also alleviate the communication overhead and simplify the IRSs deployment.

关键词： covert communications deep reinforcement learning intelligent reflecting surface

来源：评论

学校读者我要写书评

暂无评论

Knowledge Distillation on Multiple Experts for Long-Tailed Recognition

Knowledge Distillation on Multiple Experts for Long-Tailed R...

引用

2023 China Automation Congress, CAC 2023

作者： Ai, Daqing Qu, Yanyun School of Informatics Fujian Key Laboratory of Sensing and Computing for Smart City Xiamen University Fujian China

ISBN: (纸本)9798350303759

Deep learning-based image recognition encounters a great challenge since the real-world data always exhibits long-tailed distribution, that is, a few categories contain most of the examples while the rest categories have very few ones. Imbalanced distribution makes it difficult for models to achieve expected recognition performance. The works proposed to tackle this problem main focus on two strategies: re-balancing and ensemble learning. In this paper, we propose a new ensemble method: Multi-Expert Distillation for Long-tailed Recognition (MED), which consists of two training phases. Specifically, we first design a Multi-Expert Network containing multiple experts, each of which is trained in a particular domain. In this case, the whole task is split into several sub-tasks. On top of that, we perform knowledge distillation training to transfer knowledge from Multi-Expert Network to a unified student network to ensemble the experts so that the student can recognize all the categories. Besides, we construct a weight module to control the impact of each expert on the student to maximize the effectiveness of knowledge transfer. We conduct extensive experiments on three benchmarks and the results show that our proposed MED achieves compelling performance with notable improvements. © 2023 IEEE.

关键词： Image recognition

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：