检索结果-内蒙古大学图书馆

16th International Conference on Digital Image Processing, ICDIP 2024

作者： Chen, Xiuning Wang, Huabin Dai, Shulin Yuan, Hongrui Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Anhui Hefei230601 China

ISBN: (数字)9781510682917

ISBN: (纸本)9781510682900

In the domain of human pose estimation, graph convolutional networks have exhibited notable performance enhancements owing to their adeptness in naturally modeling the representation of human poses through graph structures. However, prevailing methods predominantly concentrate on the local physical connections between joints, overlooking higher-order neighboring nodes. This limitation curtails their ability to effectively exploit relationships between distant joints. This article introduces a Multiscale Spatio-Temporal Hypergraph Convolutional Network (MST-HCN) designed to capture spatio-temporal information and higher-order dependencies. MST-HCN encompasses two pivotal modules: Multiscale Hypergraph Convolution (MHCN) and Multiscale Temporal Convolution (MTCN). The MHCN module represents human poses as hypergraphs in various forms, enabling the comprehensive extraction of both local and global structural information. In contrast to traditional stride convolutions, MTCN leverages multiple branches to learn important frames based on their significance, thereby filtering out redundant frames. Experimental results underscore that MST-HCN surpasses state-of-the-art methods in benchmark tests such as Human3.6M and *** particular, our proposed MST-HCN method boosts performance by 1.5% and 0.9%, compared to the closest latest method, using detected 2D poses and ground truth 2D settings respectively. © 2024 SPIE.

关键词： Human form models

来源：评论

学校读者我要写书评

暂无评论

Enhancing Video Quality Assessment through Semantic Guidance 4

Enhancing Video Quality Assessment through Semantic Guidance

引用

4th International Symposium on Artificial Intelligence and Intelligent Manufacturing, AIIM 2024

作者： Wang, Futian Li, Zeyu School of Computer Science and Technology Anhui University Hefei China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Hefei China

ISBN: (纸本)9798331541729

With the rapid advancement of network and information technologies, user-generated content (UGC) platforms like TikTok have become essential mainstream media channels. However, video quality often varies significantly due to factors during shooting, compression, and transmission, inconveniencing users. While existing methods assess video quality using objective standards, human subjective perception remains crucial. This paper introduces a simple and effective no-reference video quality assessment model that integrates human perception into the complex video generation process to enhance UGC quality evaluation. We develop a trainable FastVQA branch to extract technical features and incorporate a ConvNeXt branch to capture semantic features. These features are combined and processed through a multi-layer perceptron (MLP) network to predict video quality scores. Our findings demonstrate that incorporating semantic features allows the network to better leverage human perception of video content quality, thereby improving the accuracy of technical assessments. This method not only optimizes traditional quality assessment models but also offers a new research direction, highlighting the importance of human perception in automated video quality evaluation. © 2024 IEEE.

关键词： Semantic Guidance UGC Video Quality Assessment

来源：评论

学校读者我要写书评

暂无评论

LiSenNet: Lightweight Sub-band and Dual-Path Modeling for Real-Time Speech Enhancement

LiSenNet: Lightweight Sub-band and Dual-Path Modeling for Re...

引用

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

作者： Yan, Haoyin Zhang, Jie Fan, Cunhang Zhou, Yeping Liu, Peiqi Hefei China Anhui Province Key Laboratory of Multimodal Cognitive Computation Anhui University Hefei China Shenzhen China

ISBN: (纸本)9798350368741

Speech enhancement (SE) aims to extract the clean waveform from noise-contaminated measurements to improve the speech quality and intelligibility. Although learning-based methods can perform much better than traditional counterparts, the large computational complexity and model size heavily limit the deployment on latency-sensitive and low-resource edge devices. In this work, we propose a lightweight SE network (LiSenNet) for real-time applications. We design sub-band downsampling and upsampling blocks and a dual-path recurrent module to capture band-aware features and time-frequency patterns, respectively. A noise detector is developed to detect noisy regions in order to perform SE adaptively and save computational costs. Compared to recent higher-resource-dependent baseline models, the proposed LiSenNet can achieve a competitive performance with only 37k parameters (half of the state-of-the-art model) and 56M multiply-accumulate (MAC) operations per second. © 2025 IEEE.

关键词： lightweight network low complexity real-time applications Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Dynamic Strip Convolution and Adaptive Morphology Perception Plugin for Medical Anatomy Segmentation

引用

IEEE Transactions on Medical Imaging 2025年第6期44卷 2541-2552页

作者： Hu, Guyue Kang, Yukun Zhao, Gangming Jin, Zhe Li, Chenglong Tang, Jin Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Security Artificial Intelligence School of Artificial Intelligence Hefei230601 China Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Hefei230601 China The University of Hong Kong Department of Computer Science Hong Kong Anhui University School of Artificial Intelligence Hefei230601 China

Medical anatomy segmentation is essential for computer-aided diagnosis and lesion localization in medical images. For example, segmenting individual ribs benefits localizing the lung lesions and providing vital medical measurements (such as rib spacing) for generating medical reports. Existing methods segment shape-different anatomies (such as striped ribs, bulky lungs, and angular scapula) with the same network architecture, the morphology heterogeneity is heavily overlooked. Although some shape-aware operators like deformable convolution and dynamic snake convolution have been introduced to cater to specific object morphology, they still struggle with orientation-varying strip structures, such as 24 ribs and 2 clavicles. In this paper, we propose a novel convolution plugin (DSC-AMP) for medical anatomy segmentation, which is comprised of a dynamic strip convolution (DSC) operator and an adaptive morphology perception (AMP) strategy. Specifically, the dynamic strip convolution customizes gradually varying directions and offsets for each local region, achieving dynamic striped receptive fields. Additionally, the adaptive morphology perception strategy incorporates insights from various shape-aware convolutional kernels, enabling the model to discern and integrate crucial representations corresponding to heterogeneous anatomies. Extensive experiments on two large-scale datasets demonstrate the effectiveness and superiority of the proposed approach for tackling heterogeneous medical anatomy segmentation. © 2025 IEEE. All rights reserved.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

CATAD: Conformer-Based Adversarial Training with Adaptive Diffusion for Bone-Conducted Speech Enhancement

CATAD: Conformer-Based Adversarial Training with Adaptive Di...

引用

International Symposium on Chinese Spoken Language Processing

作者： Zhiqiang Duan Jian Zhou Cunhang Fan Liang Tao Zhao Lv Anhui Provincial Key Laboratory of Multimodal Cognitive Computing Anhui University Hefei

ISBN: (数字)9798331516826

ISBN: (纸本)9798331516833

Due to the low-pass filtering of human tissues, the high-frequency components of bone-conducted (BC) speech suffer severe attenuation, resulting in reduced speech quality and intelligibility. To address this issue, this paper proposes a novel approach called CATAD, which employs conformer-based adversarial training with adaptive diffusion for bone-conducted speech enhancement. In CATAD, we deploy a TS-Conformer consisting of two Conformer modules that capture temporal and frequency dependencies, respectively, to enhance the expressiveness of speech features. To ensure the stability of the adversarial learning process and the diversity of the generated results, we adopt an adaptive diffusion process where random noise is added to the generated and real data, respectively. The proposed adaptive diffusion process boosts the capability of the discriminator to distinguish between real air-conducted speech and enhanced BC speech. Experimental results on the ESMB dataset demonstrate that the proposed CATAD can recover absent high components from BC speech, substantially improving speech quaility and intelligiblity of BC speech.

关键词： Training Filtering Noise Diffusion processes Speech enhancement Attenuation Generators Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

DARNet: Dual Attention Refinement Network with Spatiotemporal Construction for Auditory Attention Detection 38

DARNet: Dual Attention Refinement Network with Spatiotempora...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Yan, Sheng Fan, Cunhang Zhang, Hongyu Yang, Xiaoke Tao, Jianhua Lv, Zhao Anhui Province Key Laboratory of Multimodal Cognitive Computations School of Computer Science and Technology Anhui University China Department of Automation Tsinghua University China

At a cocktail party, humans exhibit an impressive ability to direct their attention. The auditory attention detection (AAD) approach seeks to identify the attended speaker by analyzing brain signals, such as EEG signals. However, current AAD algorithms overlook the spatial distribution information within EEG signals and lack the ability to capture long-range latent dependencies, limiting the model's ability to decode brain activity. To address these issues, this paper proposes a dual attention refinement network with spatiotemporal construction for AAD, named DARNet, which consists of the spatiotemporal construction module, dual attention refinement module, and feature fusion & classifier module. Specifically, the spatiotemporal construction module aims to construct more expressive spatiotemporal feature representations, by capturing the spatial distribution characteristics of EEG signals. The dual attention refinement module aims to extract different levels of temporal patterns in EEG signals and enhance the model's ability to capture long-range latent dependencies. The feature fusion & classifier module aims to aggregate temporal patterns and dependencies from different levels and obtain the final classification results. The experimental results indicate that DARNet achieved excellent classification performance, particularly under short decision windows. While maintaining excellent classification performance, DARNet significantly reduces the number of required parameters. Compared to the state-of-the-art models, DARNet reduces the parameter count by 91%. Code is available at: https://***/fchest/***. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-aware coarse-to-fine alignment for text-image person retrieval

引用

Visual Intelligence 2025年第1期3卷 1-14页

作者： Deng, Yifei Chen, Zhengyu Li, Chenglong Tang, Jin Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei 230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Security Artificial Intelligence School of Artificial Intelligence Anhui University Hefei 230601 China

Text-to-image person retrieval, a fine-grained cross-modal retrieval problem, aims to search for person images from an image library that match a given textual caption. Existing text-to-image person retrieval methods usually use fixed-point embedding to express the semantics of the two modalities and perform multi-granularity alignment between modalities in the embedding space. However, owing to the inherent mutual one-to-many correspondence between images and texts, it is often difficult for fixed-point embedding methods to adequately capture this relationship, leading to erroneous retrieval results. To address this problem, we propose a novel uncertainty-aware coarse-to-fine alignment method, which first maps fixed-point embedding to probability distributions and then aligns two modalities in terms of distributions and sampling points at a coarse-to-fine granularity, for accurate text-to-image person retrieval. Specifically, we first introduce two contrastive learning tasks of distribution contrast learning and point contrast learning, to achieve coarse-grained inter-modal alignment with uncertainty-aware. The distribution contrast learning task ensures that distributions with the same identity are as similar as possible across modalities through distribution-based contrastive learning. The point contrast learning task performs the contrastive learning of inter-modal and intra-modal sampling points, which not only models rich and diverse cross-modal associations, but also optimizes the learning of distributions. For the fine-grained association requirements of text-to-image person retrieval, we design the task of uncertainty-aware attribute masking language reconstruction, which achieves fine-grained alignment by randomly masking attribute words in the text and reconstructing them via inter-modal sample point interactions. Extensive experiments on two public datasets demonstrate the superior performance of our method. © The Author(s) 2025.

关键词： Coarse-to-fine alignment Contrastive learning Cross-modal retrieval Probalility distribution Uncertainty-aware

来源：评论

学校读者我要写书评

暂无评论

Non-symmetrical Sibling-Stream Network with Adaptive Positional Encoding for Automatic Medical Report Generation 5

Non-symmetrical Sibling-Stream Network with Adaptive Positio...

引用

5th International Conference on Intelligent Medicine and Image Processing, IMIP 2023

作者： Si, Jiayu Zhao, Haifeng Huang, Lili Tu, Zhengzheng Anhui University School of Computer Science and Technology Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Hefei Anhui China

ISBN: (纸本)9798350337396

Automatic medical report generation technology aims to alleviate the workload of radiologists in writing reports and avoid possible misdiagnosis and missed diagnosis. Recent solutions in medical report generation are mainly based on two mainstream architectures: convolutional neural networks (CNNs) and transformer. However, these two architectures are insufficient for the medical report generation task due to the following three reasons. First, CNN lacks the global context information due to the constrained receptive field of convolution. Second, the complete image is cut into small patches before being fed to the transformer, which leads to the damage to the local details. Third, position encoding is critical to compensate the lost position information caused by serialized input, but the conventional position encoding (e.g., absolute positional encoding) inadequately encodes position information. Therefore, we propose a novel Non-Symmetrical Sibling-Stream Network with Adaptive Positional Encoding which includes three modules: sibling-stream encoder (SSE) module, feature interaction (FI) module, and memory-driven decoder (MDD) module. In detail, in order to cope with respective drawbacks of both architectures, we first propose sibling-stream encoder (SSE) module which combines the global context by transformer and local details (e.g., texture and edge) by CNN. Concurrently, we design a new encoding method named adaptive positional encoding (APE), which can effectively encode the absolute position information within the patch and the relative position information between patches, so as to improve the accuracy of medical report. Then, the feature interaction (FI) module is designed to enhance the image feature representation by interacting the information of sibling-stream. Finally, we introduce a memory-driven decoder to implicitly model and memorize similar patterns in different medical reports during the generation process. Extensive experiments on two public data

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

Visible-infrared person re-identification via specific and shared representations learning

引用

Visual Intelligence 2023年第1期1卷 1-12页

作者： Zheng, Aihua Liu, Juncong Wang, Zi Huang, Lili Li, Chenglong Yin, Bing Information Materials and Intelligent Sensing Laboratory of Anhui Province the Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Artificial Intelligence Anhui University Hefei 230601 China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei 230601 China iFLYTEK Research No. 666 West Wangjiang Road Anhui Hefei China

The primary goal of visible-infrared person re-identification (VI-ReID) is to match pedestrian photos obtained during the day and night. The majority of existing methods simply generate auxiliary modalities to reduce the modality discrepancy for cross-modality matching. They capture modality-invariant representations but ignore the extraction of modality-specific representations that can aid in distinguishing among various identities of the same modality. To alleviate these issues, this work provides a novel specific and shared representations learning (SSRL) model for VI-ReID to learn modality-specific and modality-shared representations. We design a shared branch in SSRL to bridge the image-level gap and learn modality-shared representations, while a specific branch retains the discriminative information of visible images to learn modality-specific representations. In addition, we propose intra-class aggregation and inter-class separation learning strategies to optimize the distribution of feature embeddings at a fine-grained level. Extensive experimental results on two challenging benchmark datasets, SYSU-MM01 and RegDB, demonstrate the superior performance of SSRL over state-of-the-art methods. © The Author(s) 2023.

关键词： Cross-modality Person Re-identification (Re-ID) Shared Representations Specific Representations

来源：评论

学校读者我要写书评

暂无评论

Bilateral Masking with prompt for Knowledge Graph Completion

Bilateral Masking with prompt for Knowledge Graph Completion

引用

2024 Findings of the Association for Computational Linguistics: NAACL 2024

作者： Kong, Yonghui Fan, Cunhang Chen, Yujie Zhang, Shuai Lv, Zhao Tao, Jianhua Anhui Province Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China Department of Automation Tsinghua University China

ISBN: (纸本)9798891761193

The pre-trained language model (PLM) has achieved significant success in the field of knowledge graph completion (KGC) by effectively modeling entity and relation descriptions. In recent studies, the research in this field has been categorized into methods based on word matching and sentence matching, with the former significantly lags behind. However, there is a critical issue in word matching methods, which is that these methods fail to obtain satisfactory single embedding representations for *** address this issue and enhance entity representation, we propose the Bilateral Masking with prompt for Knowledge Graph Completion (BMKGC) *** methodology employs prompts to narrow the distance between the predicted entity and the known entity. Additionally, the BMKGC model incorporates a bi-encoder architecture, enabling simultaneous predictions at both the head and tail. Furthermore, we propose a straightforward technique to augment positive samples, mitigating the problem of degree bias present in knowledge graphs and thereby improving the model's robustness. Experimental results conclusively demonstrate that BMKGC achieves state-of-the-art performance on the WN18RR dataset. © 2024 Association for Computational Linguistics.

关键词： Knowledge graph

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：