检索结果-南通市图书馆

Clustering Reference Images Based on Covisibility for Visual Localization

Computers, Materials & Continua 2023年第5期75卷 2705-2725页

作者： Sangyun Lee Junekoo Kang hyunki hong School of Integrative Engineering Chung-Ang UniversityHeukseok-ro 84Dongjak-kuSeoul06973Korea School of Software Chung-Ang UniversityHeukseok-ro 84Dongjak-kuSeoul06973Korea

In feature-based visual localization for small-scale scenes,local descriptors are used to estimate the camera pose of a query *** large and ambiguous environments,learning-based hierarchical networks that employ local as well as global descriptors to reduce the search space of database images into a smaller set of reference views have been ***,since global descriptors are generated using visual features,reference images with some of these features may be erroneously *** order to address this limitation,this paper proposes two clustering methods based on how often features appear as well as their *** both approaches,the scene is represented by voxels whose size and number are computed according to the size of the scene and the number of available *** the first approach,a voxel-based histogram representing highly reoccurring scene regions is generated from reference images.A meanshift is then employed to group the most highly reoccurring voxels into place clusters based on their spatial *** the second approach,a graph representing the covisibility-based relationship of voxels is *** matching is performed within the reference image clusters,and a perspective-n-point is employed to estimate the camera *** experimental results showed that camera pose estimation using the proposed approaches was more accurate than that of previous methods.

关键词： Visual localization deep learning voxel representation clustering covisibility meanshift graph structure

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Use of Local Region Maps on Convolutional LSTM for Single-Image HDR Reconstruction

引用

Computers, Materials & Continua 2022年第6期71卷 4555-4572页

作者： Seungwook Oh GyeongIk Shin hyunki hong College of Software Chung-Ang UniversityHeukseok-ro 84Dongjak-kuSeoul06973Korea

Low dynamic range(LDR)images captured by consumer cameras have a limited luminance *** the conventional method for generating high dynamic range(HDR)images involves merging multiple-exposure LDR images of the same scene(assuming a stationary scene),we introduce a learning-based model for single-image HDR *** input LDR image is sequentially segmented into the local region maps based on the cumulative histogram of the input brightness *** the local region maps,SParam-Net estimates the parameters of an inverse tone mapping function to generate a pseudo-HDR *** process the segmented region maps as the input sequences on long short-term ***,a fast super-resolution convolutional neural network is used for HDR image *** proposed method was trained and tested on datasets including HDR-Real,LDR-HDR-pair,and *** experimental results revealed that HDR images can be generated more reliably than using contemporary end-to-end approaches.

关键词： Low dynamic range high dynamic range deep learning convolutional long short-term memory inverse tone mapping function

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Using Speaker-Specific Emotion Representations in Wav2vec 2.0-Based Modules for Speech Emotion Recognition

引用

Computers, Materials & Continua 2023年第10期77卷 1009-1030页

作者： Somin Park Mpabulungi Mark Bogyung Park hyunki hong College of Software Chung-Ang UniversitySeoul06973Korea Department of AI Chung-Ang UniversitySeoul06973Korea

Speech emotion recognition is essential for frictionless human-machine interaction,where machines respond to human instructions with context-aware *** properties of individuals’voices vary with culture,language,gender,and *** variations in speaker-specific properties may hamper the performance of standard representations in downstream tasks such as speech emotion recognition(SER).This study demonstrates the significance of speaker-specific speech characteristics and how considering them can be leveraged to improve the performance of SER *** the proposed approach,two wav2vec-based modules(a speaker-identification network and an emotion classification network)are trained with the Arcface *** speaker-identification network has a single attention block to encode an input audio waveform into a speaker-specific *** emotion classification network uses a wav2vec 2.0-backbone as well as four attention blocks to encode the same input audio waveform into an emotion *** two representations are then fused into a single vector representation containing emotion and speaker-specific *** results showed that the use of speaker-specific characteristics improves SER ***,combining these with an angular marginal loss such as the Arcface loss improves intra-class compactness while increasing inter-class separability,as demonstrated by the plots of t-distributed stochastic neighbor embeddings(t-SNE).The proposed approach outperforms previous methods using similar training strategies,with a weighted accuracy(WA)of 72.14%and unweighted accuracy(UA)of 72.97%on the Interactive Emotional Dynamic Motion Capture(IEMOCAP)*** demonstrates its effectiveness and potential to enhance human-machine interaction through more accurate emotion recognition in speech.

关键词： Attention block IEMOCAP dataset speaker-specific representation speech emotion recognition wav2vec 2.0

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

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：