检索结果-内蒙古大学图书馆

arXiv 2025年

作者： He, Yuwen Wang, Wei Wang, Wanyu Jiang, Kui Wu, Wanyu Computer Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology China School of Computer Science and Technology Harbin Institute of Technology China

Lens flares arise from light reflection and refraction within sensor arrays, whose diverse types include glow, veiling glare, reflective flare and so on. Existing methods are specialized for one specific type only, and overlook the simultaneous occurrence of multiple typed lens flares, which is common in the real-world, e.g. coexistence of glow and displacement reflections from the same light source. These co-occurring lens flares cannot be effectively resolved by the simple combination of individual flare removal methods, since these coexisting flares originates from the same light source and are generated simultaneously within the same sensor array, exhibit a complex interdependence rather than simple additive relation. To model this interdependent flares’ relationship, our Nighttime Lens Flare Formation model is the first attempt to learn the intrinsic physical relationship between flares on the imaging plane. Building on this physical model, we introduce a solution to this joint flare removal task named Self-supervised Generation-based Lens Flare Removal Network (SGLFR-Net), which is self-supervised without pre-training. Specifically, the nighttime glow is detangled in PSF Rendering Network(PSFR-Net) based on PSF Rendering Prior, while the reflective flare is modelled in Texture Prior Based Reflection Flare Removal Network (TPRR-Net). Empirical evaluations demonstrate the effectiveness of the proposed method in both joint and individual glare removal tasks. Copyright © 2025, The Authors. All rights reserved.

关键词： Glare

来源：评论

学校读者我要写书评

暂无评论

Toward Comprehensive Semantic Prompt for Region Contrastive Learning Underwater Image Enhancement

Toward Comprehensive Semantic Prompt for Region Contrastive ...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Xiao Wang Yongsheng Fu Wei Wang Wei Liu School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Underwater image enhancement (UIE) focuses on mitigating image quality degradation due to light absorption and scattering. However, most existing methods enhance images via a global and uniform manner, neglecting the inherent semantic information in different regions, which may cause the network to easily deviate from the region’s original color. Moreover, these methods typically rely on clear images to guide network convergence, a process constrained by the limited availability of real-world datasets, making it extremely challenging to train enhancement models for various degradations. To address these challenges, this paper introduces a semantic guidance and region contrastive constraints network (SRCNet). Initially, we propose a semantic-aware RWKV (Receptance Weighted key Value) block and a semantic prompt regularization module. These components leverage intra-target semantic correlations to preserve image details and colors within a global perceptual framework, while employing focal loss to emphasize the restoration of severely degraded regions. Subsequently, we introduce a region contrastive learning method that effectively utilizes negative samples to precisely capture features sensitive to degradation factors, thereby fostering robust feature distributions. Finally, experimental results demonstrate that our method outperforms existing state-of-the-art (SOTA) approaches.

关键词： Degradation Visualization Image color analysis Semantics Signal processing algorithms Contrastive learning Speech enhancement Signal processing Image restoration Image enhancement

来源：评论

学校读者我要写书评

暂无评论

VAGeo: View-specific Attention for Cross-View Object Geo-Localization

arXiv

引用

arXiv 2025年

作者： Li, Zhongyang Yuan, Xin Liu, Wei Xu, Xin School of Computer Science and Technology Wuhan University of Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System China

Cross-view object geo-localization (CVOGL) aims to locate an object of interest in a captured ground- or drone-view image within the satellite image. However, existing works treat ground-view and drone-view query images equivalently, overlooking their inherent viewpoint discrepancies and the spatial correlation between the query image and the satellite-view reference image. To this end, this paper proposes a novel View-specific Attention Geo-localization method (VAGeo) for accurate CVOGL. Specifically, VAGeo contains two key modules: view-specific positional encoding (VSPE) module and channel-spatial hybrid attention (CSHA) module. In object-level, according to the characteristics of different viewpoints of ground and drone query images, viewpoint-specific positional codings are designed to more accurately identify the click-point object of the query image in the VSPE module. In feature-level, a hybrid attention in the CSHA module is introduced by combining channel attention and spatial attention mechanisms simultaneously for learning discriminative features. Extensive experimental results demonstrate that the proposed VAGeo gains a significant performance improvement, i.e., improving acc@0.25/acc@0.5 on the CVOGL dataset from 45.43%/42.24% to 48.21%/45.22% for ground-view, and from 61.97%/57.66% to 66.19%/61.87% for drone-view. Copyright © 2025, The Authors. All rights reserved.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

Minimizing Disparities between Real and Pseudo Queries for Unsupervised Visual Grounding

Minimizing Disparities between Real and Pseudo Queries for U...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Hui Jiang Changkai Ji Jilan Xu Yanhao Zhu Yuejie Zhang Rui Feng Tao Zhang Shang Gao School of Computer Science Shanghai Key Laboratory of Intelligent Information Processing Fudan University Shanghai China School of Information Management and Engineering Shanghai Key Laboratory of Financial Information Technology Shanghai University of Finance and Economics Shanghai China School of Information Technology Deakin University Victoria Australia

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Visual grounding involves the identification and localization of image regions given textual descriptions. To reduce the manual labeling effort on region-text pairs, unsupervised visual grounding aims to generate pseudo bounding box and query pairs for training grounding models. However, there exists significant disparities between real and pseudo queries in terms of object, attribute distributions, and textual formats, limiting the generalization performance of unsupervised grounding methods. To address this challenge, we propose a novel unsupervised visual grounding framework. During training, we prompt Multimodal Large Language Models to generate pseudo queries, in which the entities are beyond the object detector’s pre-defined limited categories, and are associated with richer attributes. We further devise a Modifier Tree structure to bridge the gap of textual format between real and pseudo queries. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art unsupervised approaches on public benchmark datasets, particularly when dealing with complex queries.

关键词： Training Location awareness Visualization Limiting Grounding Large language models Manuals Labeling Speech processing Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Event-based Video Person Re-identification via Cross-Modality and Temporal Collaboration

arXiv

引用

arXiv 2025年

作者： Li, Renkai Yuan, Xin Liu, Wei Xu, Xin School of Computer Science and Technology Wuhan University of Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System China

Video-based person re-identification (ReID) has become increasingly important due to its applications in video surveillance applications. By employing events in video-based person ReID, more motion information can be provided between continuous frames to improve recognition accuracy. Previous approaches have assisted by introducing event data into the video person ReID task, but they still cannot avoid the privacy leakage problem caused by RGB images. In order to avoid privacy attacks and to take advantage of the benefits of event data, we consider using only event data. To make full use of the information in the event stream, we propose a Cross-Modality and Temporal Collaboration (CMTC) network for event-based video person ReID. First, we design an event transform network to obtain corresponding auxiliary information from the input of raw events. Additionally, we propose a differential modality collaboration module to balance the roles of events and auxiliaries to achieve complementary effects. Furthermore, we introduce a temporal collaboration module to exploit motion information and appearance cues. Experimental results demonstrate that our method outperforms others in the task of event-based video person ReID. Copyright © 2025, The Authors. All rights reserved.

关键词： information leakage

来源：评论

学校读者我要写书评

暂无评论

A High Performance Region Recognition Network-Enhanced Deep CNN for Image Content Perceptual Hashing

引用

IEEE Internet of Things Journal 2025年

作者： Yang, Meihong Qi, Baolin Ma, Bin Xu, Jian Xian, Yongjin Li, Xiaolong Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China School of Computer and Information Engineering Shandong University of Finance and Economics Jinan China Beijing Jiaotong University School of Computer and Information Technology Beijing100044 China

Perceptual image hashing has emerged as a crucial forensic tool within the Internet of Things (IoT) ecosystem. Traditional perceptual hashing algorithms predominantly rely on global image features to generate hash codes, which limit their ability to represent key features of images effectively. This paper introduces a Perceptual Region Recognition Network (PRRN) to accurately identify key feature regions in images based on their texture distribution characteristics, thereby generating image perceptual hashing codes that reflect the key content of the images. At the same time, a perceptual hashing feature extraction module, which integrates a Residual Network (ResNet) and a Weighted Feature Fusion Network (WFFN), is built to extract deep semantic features of the object image. Where, ResNet is leveraged to extract high-level semantic features, while WFFN ensures the preservation of low-level local features. Furthermore, skip connections are employed to achieve content enhancements for intricate details of critical image regions. Additionally, the Mean Squared Error (MSE) loss is incorporated to enhance the accuracy of key region localization, further improving the sensitivity of image perceptual hash codes and accelerating the network’s convergence speed. Extensive experimental evaluations demonstrate that the proposed PRRN-based perceptual image hashing scheme significantly outperforms other state-of-the-art methods in terms of image feature representation capability. Specifically, it achieves an average improvement of over 1.2% in attack-resistant capability for images compared with other counterparts, making it a promising candidate for practical applications in the IoT environment. © 2014 IEEE.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

Stable Extended U-Net for Noise-Robust Speaker Verification

Stable Extended U-Net for Noise-Robust Speaker Verification

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal processing, ICASSP 2025

作者： Wang, Zonghui Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China Department of Electronic Engineering Tsinghua University Beijing100084 China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)9798350368741

With advancements in deep learning, speaker verification systems have significantly improved their performance in noisy environments. Researchers typically demonstrate the effectiveness of their improved models by comparing performance on specific datasets, such as the VoxCeleb benchmark. However, in diverse real-world noise conditions, the out-of-domain generalization ability is also a crucial factor in evaluating a model's performance improvement. Research on stable learning indicates that eliminating the spurious correlation between training and testing data can enhance the generalization of the model. Building on this idea, we propose an improved speaker verification system with high generalization based on the extended U-Net (ExU-Net). It uses the sample reweighting method from stable learning to eliminate sample correlations and retains more effective speaker information through sub-pixel convolutions and coordinate attention mechanisms. We validate the effectiveness of this approach through extensive evaluations on VoxCeleb1, VOiCES, and other out-of-domain noise test sets, highlighting its generalization capability and model robustness. © 2025 IEEE.

关键词： noisy environments out-of-domain generalization Speaker verification stable learning

来源：评论

学校读者我要写书评

暂无评论

Resource Allocation for MEC in Ultra-dense Networks

引用

Journal of computers (Taiwan) 2025年第1期36卷 143-162页

作者： Ma, Huan Li, Shuyun Wang, Zhihui Dou, Xubing Zhang, Xinchang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Qilu University of Technology Shandong Academy of Sciences Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan China Shandong Normal University Jinan China

In the complex context of 5G ultra-dense networks, the unprecedented growth in traffic and computing demand has made the interaction between edge devices and edge servers extremely intricate due to the time-varying nature of the network environment and resource heterogeneity. The combination of UDN and Mobile Edge Computing not only significantly improves network performance but also expands application scenarios and enhances user experience, marking it as one of the crucial trends in future network development. In this paper, we review recent results and research progress on MEC-based resource allocation in 5G ultra-dense networks. We analyze resource allocation strategies from multiple dimensions, including user association, interference management, energy efficiency optimization, delay reduction, access selection, and user mobility. Finally, we discuss the challenges and open issues that need to be addressed for future development in this area. © 2025 Codon Publications. All rights reserved.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Anomize: Better Open Vocabulary Video Anomaly Detection

arXiv

引用

arXiv 2025年

作者： Li, Fei Liu, Wenxuan Chen, Jingjing Zhang, Ruixu Wang, Yuran Zhong, Xian Wang, Zheng National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China State Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China Hubei Key Laboratory of Transportation Internet of Things Wuhan University of Technology China

Open Vocabulary Video Anomaly Detection (OVVAD) seeks to detect and classify both base and novel anomalies. However, existing methods face two specific challenges related to novel anomalies. The first challenge is detection ambiguity, where the model struggles to assign accurate anomaly scores to unfamiliar anomalies. The second challenge is categorization confusion, where novel anomalies are often misclassified as visually similar base instances. To address these challenges, we explore supplementary information from multiple sources to mitigate detection ambiguity by leveraging multiple levels of visual data alongside matching textual information. Furthermore, we propose incorporating label relations to guide the encoding of new labels, thereby improving alignment between novel videos and their corresponding labels, which helps reduce categorization confusion. The resulting Anomize framework effectively tackles these issues, achieving superior performance on UCF-CRIME and XD-VIOLENCE datasets, demonstrating its effectiveness in OVVAD. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MEFA-Net: A mask enhanced feature aggregation network for polyp segmentation

引用

computers in Biology and Medicine 2025年 186卷 109601-109601页

作者： Ke, Xiao Chen, Guanhong Liu, Hao Guo, Wenzhong Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fuzhou350116 China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou350116 China

Accurate polyp segmentation is crucial for early diagnosis and treatment of colorectal cancer. This is a challenging task for three main reasons: (i) the problem of model overfitting and weak generalization due to the multi-center distribution of data;(ii) the problem of interclass ambiguity caused by motion blur and overexposure to endoscopic light;and (iii) the problem of intraclass inconsistency caused by the variety of morphologies and sizes of the same type of polyps. To address these challenges, we propose a new high-precision polyp segmentation framework, MEFA-Net, which consists of three modules, including the plug-and-play Mask Enhancement Module (MEG), Separable Path Attention Enhancement Module (SPAE), and Dynamic Global Attention Pool Module (DGAP). Specifically, firstly, the MEG module regionally masks the high-energy regions of the environment and polyps through a mask, which guides the model to rely on only a small amount of information to distinguish between polyps and background features, avoiding the model from overfitting the environmental information, and improving the robustness of the model. At the same time, this module can effectively counteract the "dark corner phenomenon" in the dataset and further improve the generalization performance of the model. Next, the SPAE module can effectively alleviate the inter-class fuzzy problem by strengthening the feature expression. Then, the DGAP module solves the intra-class inconsistency problem by extracting the invariance of scale, shape and position. Finally, we propose a new evaluation metric, MultiColoScore, for comprehensively evaluating the segmentation performance of the model on five datasets with different domains. We evaluated the new method quantitatively and qualitatively on five datasets using four metrics. Experimental results show that MEFA-Net significantly improves the accuracy of polyp segmentation and outperforms current state-of-the-art algorithms. Code posted on https://***/

关键词： Endoscopy

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：