检索结果-内蒙古大学图书馆

Explaining the better generalization of label distribution learning for classification

Science China(Information Sciences) 2025年第5期68卷 348-361页

作者： Jing WANG Xin GENG School of Computer Science and Engineering Southeast University Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications(Southeast University) Ministry of Education

Label distribution learning(LDL) has shown advantages over traditional single-label learning(SLL) in many realworld applications, but its superiority has not been theoretically understood. In this paper, we attempt to explain why LDL generalizes better than SLL. Label distribution has rich supervision information such that an LDL method can still choose the sub-optimal label from label distribution even if it neglects the optimal one. In comparison, an SLL method has no information to choose from when it fails to predict the optimal label. The better generalization of LDL can be credited to the rich information of label distribution. We further establish the label distribution margin theory to prove this explanation; inspired by the theory,we put forward a novel LDL approach called LDL-LDML. In the experiments, the LDL baselines outperform the SLL ones, and LDL-LDML achieves competitive performance against existing LDL methods, which support our explanation and theories in this paper.

关键词： label distribution learning classification generalization label distribution margin label ambiguity neural network

来源：评论

学校读者我要写书评

暂无评论

Localization of Autonomous Vehicles in Tunnels Based on Roadside Multi-sensor Fusion

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-13页

作者： Zhu, Kongtao Chen, Shitao Shi, Jiamin Zhu, Ziyu Zheng, Nanning National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China

Tunnels present significant challenges for the navigation and localization of autonomous vehicles due to the lack of GNSS signals and the presence of uniform scene textures. Current cooperative positioning strategies, which rely on RSS and ToF, are less effective in tunnel environments due to the unique electromagnetic conditions. To address this issue, a novel Vehicle-to-Infrastructure (V2I) cooperative localization methodology is introduced. The system comprises a roadside subsystem and a vehicle-side subsystem. In the roadside subsystem, data from multiple sensors is collected to calculate the vehicleâ's position. This information is then transmitted to the vehicle-side subsystem via V2I communication, where it is fused with onboard module data. A highly effective co-location process for the vehicle-side subsystem is proposed, along with an exemplary algorithm. This process resolves issues related to data dimension inconsistency, delay, and pose fusion commonly encountered during co-location. Furthermore, a comprehensive investigation and analysis explores factors that may impact positioning performance. This meticulous examination enhances the understanding of the framework and reveals its limitations. IEEE

关键词： Location

来源：评论

学校读者我要写书评

暂无评论

Perpendicular-Cutdepth:Perpendicular Direction Depth Cutting data Augmentation Method

引用

Computers, Materials & Continua 2024年第4期79卷 927-941页

作者： Le Zou Linsong Hu Yifan Wang Zhize Wu Xiaofeng Wang Anhui Provincial Engineering Laboratory of Big Data Technology Application forUrban Infrastructure School ofArtificial Intelligence and Big DataHefei UniversityHefei230601China Institute of Applied Optimization School of Artificial Intelligence and Big DataHefei UniversityHefei230601China

Depth estimation is an important task in computer *** data at scale for monocular depth estimation is challenging,as this task requires simultaneously capturing RGB images and depth ***,data augmentation is crucial for this *** data augmentationmethods often employ pixel-wise transformations,whichmay inadvertently disrupt edge *** this paper,we propose a data augmentationmethod formonocular depth estimation,which we refer to as the Perpendicular-Cutdepth *** method involves cutting realworld depth maps along perpendicular directions and pasting them onto input images,thereby diversifying the data without compromising edge *** validate the effectiveness of the algorithm,we compared it with existing convolutional neural network(CNN)against the current mainstream data augmentation ***,to verify the algorithm’s applicability to Transformer networks,we designed an encoder-decoder network structure based on Transformer to assess the generalization of our proposed *** results demonstrate that,in the field of monocular depth estimation,our proposed method,Perpendicular-Cutdepth,outperforms traditional data *** the indoor dataset NYU,our method increases accuracy from0.900 to 0.907 and reduces the error rate from0.357 to *** the outdoor dataset KITTI,our method improves accuracy from 0.9638 to 0.9642 and decreases the error rate from 0.060 to 0.0598.

关键词： Perpendicular depth estimation data augmentation

来源：评论

学校读者我要写书评

暂无评论

BatStyler: Advancing Multi-category Style Generation for Source-free Domain Generalization

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 5419-5430页

作者： Xu, Xiusheng Qi, Lei Zhou, Jingyang Geng, Xin Southeast University School of Computer Science and Engineering China Interdisciplinary Applications Southeast University Key Laboratory of New Generation Artificial Intelligence Technology Ministry of Education 211189 China

Source-Free Domain Generalization (SFDG) aims to develop a model that performs on unseen domains without relying on any source domains. However, the implementation remains constrained due to the unavailability of training data. Research on SFDG focus on knowledge transfer of multi-modal models and style synthesis based on joint space of multiple modalities, thus eliminating the dependency on source domain images. However, existing works primarily work for multi-domain and less-category configuration, but performance on multi-domain and multi-category configuration is relatively poor. In addition, the efficiency of style synthesis also deteriorates in multi-category scenarios. How to efficiently synthesize sufficiently diverse data and apply it to multi-category configuration is a direction with greater practical value. In this paper, we propose a method called BatStyler, which is utilized to improve the capability of style synthesis in multi-category scenarios. BatStyler consists of two modules: Coarse Semantic Generation and Uniform Style Generation modules. The Coarse Semantic Generation module extracts coarse-grained semantics to prevent the compression of space for style diversity learning in multi-category configuration, while the Uniform Style Generation module provides a template of styles that are uniformly distributed in space and implements parallel training. Extensive experiments demonstrate that our method exhibits comparable performance on less-category datasets, while surpassing state-of-the-art methods on multi-category datasets. Code is available here. © 1991-2012 IEEE.

关键词： Knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Event-enhanced synthetic aperture imaging

引用

Science China(Information Sciences) 2025年第3期68卷 388-390页

作者： Siqi LI Shaoyi DU Jun-Hai YONG Yue GAO Beijing National Research Center for Information Science and Technology School of SoftwareTsinghua University Institute for Brain and Cognitive Sciences Beijing Laboratory of Brain and Cognitive IntelligenceTsinghua University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and ApplicationsInstitute of Artificial Intelligence and Robotics Xi'an Jiaotong University

Synthetic aperture imaging(SAI) methods aim to see through dense occlusions and reconstruct the target scene behind occlusions. Traditional frame-based SAI methods,e.g., DeOccNet [1], take the occluded light field ima...

关键词：

来源：评论

学校读者我要写书评

暂无评论

Visual-Linguistic Feature Alignment With Semantic and Kinematic Guidance for Referring Multi-Object Tracking

引用

IEEE Transactions on Multimedia 2025年 27卷 3034-3044页

作者： Li, Yizhe Zhou, Sanping Qin, Zheng Wang, Le Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Xi'an710049 China Xi'an Jiaotong University Institute of Artificial Intelligence and Robotics Xi'an710049 China

Referring Multi-Object Tracking (RMOT) aims to dynamically track an arbitrary number of referred targets in a video sequence according to the language expression. Previous methods mainly focus on cross-modal fusion at the feature level with designed structures. However, the insufficient visual-linguistic alignment is prone to causing visual-linguistic mismatches, leading to some targets being tracked but not correctly referred especially when facing the language expression with complex semantics or motion descriptions. To this end, we propose to conduct visual-linguistic alignment with semantic and kinematic guidance to effectively align the visual features with more diverse language expressions. In this paper, we put forward a novel end-to-end RMOT framework SKTrack, which follows the transformer-based architecture with a Language-Guided Decoder (LGD) and a Motion-Aware Aggregator (MAA). In particular, the LGD performs deep semantic interaction layer-by-layer in a single frame to enhance the alignment ability of the model, while the MAA conducts temporal feature fusion and alignment across multiple frames to enable the alignment between visual targets and language expression with motion descriptions. Extensive experiments on the Refer-KITTI and Refer-KITTI-v2 demonstrate that SKTrack achieves state-of-the-art performance and verify the effectiveness of our framework and its components. © 1999-2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Transverse Velocity Field Measurement of Solar High-resolution Images Based on Unsupervised Deep Learning

引用

Research in Astronomy and Astrophysics 2025年第3期25卷 236-248页

作者： Zhen-Hong Shang Long Chen Zhen-Ping Qiang Yi Bi Run-Xin Li Faculty of Information Engineering and Automation Kunming University of Science and Technology Yunnan Key Laboratory of Artificial Intelligence Kunming University of Science and Technology College of Big Data and Intelligent Engineering Southwest Forestry University Yunnan Observatories Chinese Academy of Sciences

Measuring the transverse velocity field in high-resolution solar images is essential for understanding solar *** paper introduces an innovative unsupervised deep learning optical flow model designed to calculate the transverse velocity field,addressing the challenges of missing optical flow labels and the limited accuracy of velocity field measurements in high-resolution solar *** proposed method converts the transverse velocity field computation problem into an optical flow computation problem,using two forward propagations of features to get rid of the reliance on optical flow ***,it reduces the impact of the“Brightness Consistency”constraint on optical flow accuracy by identifying and handling optical flow *** apply this method to compute the transverse velocity fields of high-resolution solar image sequences from the Hαand TiO bands,observed by the New Vacuum Solar *** experiments with several wellestablished optical flow methods,including those based on supervised deep learning models,show that our approach outperforms the comparison methods according to key evaluation metrics such as Residual Map Mean,Residual Map Variance,Cross Correlation,and Structural Similarity Index ***,since optical flow captures the fundamental motion information in image sequences,the proposed method can be applied to a variety of research areas,including solar image registration,sequence alignment,image super-resolution,magnetic field calibration,and solar activity *** code is available at https://***/jackie-willianm/Transverse-Velocity-Field-Measurement-of-Solar-High-Resolution-Images.

关键词： methods: data analysis techniques: image processing Sun: fundamental parameters

来源：评论

学校读者我要写书评

暂无评论

Human-Like Decision-Making of Autonomous Vehicles in Dynamic Traffic Scenarios

引用

IEEE/CAA Journal of Automatica Sinica 2023年第10期10卷 1905-1917页

作者： Tangyike Zhang Junxiang Zhan Jiamin Shi Jingmin Xin Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and ApplicationsXi’an 710049 Institute of Artificial Intelligence and Robotics Xi’an Jiaotong UniversityXi’an 710049China Momenta Shanghai 201804China

With the maturation of autonomous driving technology, the use of autonomous vehicles in a socially acceptable manner has become a growing demand of the public. Human-like autonomous driving is expected due to the impact of the differences between autonomous vehicles and human drivers on *** human-like decision-making has become a research hotspot, a unified theory has not yet been formed, and there are significant differences in the implementation and performance of existing methods. This paper provides a comprehensive overview of human-like decision-making for autonomous vehicles. The following issues are discussed: 1) The intelligence level of most autonomous driving decision-making algorithms;2) The driving datasets and simulation platforms for testing and verifying human-like decision-making;3) The evaluation metrics of human-likeness;personalized driving;the application of decisionmaking in real traffic scenarios;and 4) The potential research direction of human-like driving. These research results are significant for creating interpretable human-like driving models and applying them in dynamic traffic scenarios. In the future, the combination of intuitive logical reasoning and hierarchical structure will be an important topic for further research. It is expected to meet the needs of human-like driving.

关键词： Autonomous vehicles decision-making driving behavior human-like driving

来源：评论

学校读者我要写书评

暂无评论

ViGT: proposal-free video grounding with a learnable token in the transformer

引用

Science China(Information Sciences) 2023年第10期66卷 196-212页

作者： Kun LI Dan GUO Meng WANG School of Computer Science and Information Engineering Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Intelligent Interconnected Systems Laboratory of Anhui Province Institute of Artificial Intelligence Hefei Comprehensive National Science Center

The video grounding(VG) task aims to locate the queried action or event in an untrimmed video based on rich linguistic descriptions. Existing proposal-free methods are trapped in the complex interaction between video and query, overemphasizing cross-modal feature fusion and feature correlation for VG. In this paper, we propose a novel boundary regression paradigm that performs regression token learning in a transformer. Particularly, we present a simple but effective proposal-free framework, namely video grounding transformer(ViGT), which predicts the temporal boundary using a learnable regression token rather than multi-modal or cross-modal features. In ViGT, the benefits of a learnable token are manifested as follows.(1) The token is unrelated to the video or the query and avoids data bias toward the original video and query.(2) The token simultaneously performs global context aggregation from video and query ***, we employed a sharing feature encoder to project both video and query into a joint feature space before performing cross-modal co-attention(i.e., video-to-query attention and query-to-video attention) to highlight discriminative features in each modality. Furthermore, we concatenated a learnable regression token [REG] with the video and query features as the input of a vision-language transformer. Finally, we utilized the token [REG] to predict the target moment and visual features to constrain the foreground and background probabilities at each timestamp. The proposed ViGT performed well on three public datasets:ANet-Captions, TACoS, and YouCookⅡ. Extensive ablation studies and qualitative analysis further validated the interpretability of ViGT.

关键词： video grounding temporal sentence grounding boundary regression token learning proposal-free

来源：评论

学校读者我要写书评

暂无评论

A Method for Fast Feature Selection Utilizing Cross-Similarity within the Context of Fuzzy Relations

引用

Computers, Materials & Continua 2025年第4期83卷 1195-1218页

作者： Wenchang Yu Xiaoqin Ma Zheqing Zhang Qinli Zhang School of Big Data and Artificial Intelligence Chizhou UniversityChizhou247000China Anhui Education Big Data Intelligent Perception and Application Engineering Research Center Anhui Provincial Joint Construction Key Laboratory of Intelligent Education Equipment and TechnologyChizhou247000China

Feature selection methods rooted in rough sets confront two notable limitations:their high computa-tional complexity and sensitivity to noise,rendering them impractical for managing large-scale and noisy *** primary issue stems from these methods’undue reliance on all *** overcome these challenges,we introduce the concept of cross-similarity grounded in a robust fuzzy relation and design a rapid and robust feature selection ***,we construct a robust fuzzy relation by introducing a truncation ***,based on this fuzzy relation,we propose the concept of cross-similarity,which emphasizes the sample-to-sample similarity relations that uniquely determine feature importance,rather than considering all such relations *** studying the manifestations and properties of cross-similarity across different fuzzy granularities,we propose a forward greedy feature selection algorithm that leverages cross-similarity as the foundation for information *** algorithm significantly reduces the time complexity from O(m2n2)to O(mn2).Experimental findings reveal that the average runtime of five state-of-the-art comparison algorithms is roughly 3.7 times longer than our algorithm,while our algorithm achieves an average accuracy that surpasses those of the five comparison algorithms by approximately 3.52%.This underscores the effectiveness of our *** paper paves the way for applying feature selection algorithms grounded in fuzzy rough sets to large-scale gene datasets.

关键词： Fuzzy rough sets feature selection cross-similarity fuzzy relations

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：