检索结果-内蒙古大学图书馆

Hybrid mesh-neural representation for 3D transparent object reconstruction

Computational Visual Media 2025年第1期11卷 123-140页

作者： Jiamin Xu Zihan Zhu Hujun Bao Weiwei Xu College of Computer Science Hangzhou Dianzi UniversityHangzhou 310018China State Key Lab of CAD&CG Zhejiang UniversityHangzhou 310058China College of Computer Science ETH ZurichZurich 8092Switzerland.

In this study,we propose a novel method to reconstruct the 3D shapes of transparent objects using images captured by handheld cameras under natural lighting *** combines the advantages of an explicit mesh and multi-layer perceptron(MLP)network as a hybrid representation to simplify the capture settings used in recent *** obtaining an initial shape through multi-view silhouettes,we introduced surface-based local MLPs to encode the vertex displacement field(VDF)for reconstructing surface *** design of local MLPs allowed representation of the VDF in a piecewise manner using two-layer MLP networks to support the optimization *** local MLPs on the surface instead of on the volume also reduced the search *** a hybrid representation enabled us to relax the ray–pixel correspondences that represent the light path constraint to our designed ray–cell correspondences,which significantly simplified the implementation of a single-image-based environment-matting *** evaluated our representation and reconstruction algorithm on several transparent objects based on ground truth *** experimental results show that our method produces high-quality reconstructions that are superior to those of state-of-the-art methods using a simplified data-acquisition setup.

关键词： transparent object 3D reconstruction environment matting neural rendering

来源：评论

学校读者我要写书评

暂无评论

Facial Sentiment Analysis Using Convolutional Neural Network and Fuzzy Systems

Informatica (Slovenia)

引用

Informatica (Slovenia) 2024年第12期48卷 15-32页

作者： Kadhim, Ahmed R. Khudeyer, Raidah S. Alabbas, Maytham Department of Computer Information System College of Computer Science and Information Technology University of Basrah Basrah Iraq Department of Computer Science College of Computer Science and Information Technology University of Basrah Basrah Iraq

This study provides a detailed study of a Сonvolutional Neural Network (СNN) model optimized for facial eхpression recognition with Fuzzy logic using Fuzzy2DPooling and Fuzzy Neural Networks (FNN), and discusses data augmentation in model optimization. It highlights important roles. performance. First, the effectiveness of the models in classifying emotions from FER2013, RAB-DB, and СK+ datasets was evaluated by a 5-fold cross-validation method, which showed that the accuracy varied widely among different emotion classes and was affected by overfitting. It turned out to be easy. The integration of data augmentation techniques, including random rotation, translation, and inversion, significantly improved the model's generalization capabilities. This was evidenced by higher accuracy and more consistent loss curves observed across all folds. After augmentation, the model showed significant improvement, achieving average test accuracies of 98.95% on FER2013, 99.99% on RAF-DB, and 100% on СK+ across all folds. Despite these advances, challenges specific to certain classes of emotions remain, highlighting the need for continued model refinement. This study concludes that data augmentation is an important step in developing robust facial eхpression recognition models and has potential benefits for a variety of applications requiring accurate emotion recognition. © 2024 Slovene Society Informatika. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A SITS-Net-Based intelligent identification model for lizards 4

A SITS-Net-Based intelligent identification model for lizard...

引用

4rth International Conference on Applied Mathematics, Modelling, and Intelligent Computing, CAMMIC 2024

作者： Jiang, Yilin Ying, Zuocan Yin, Xuanyu Guo, Haidong College of Educational Science and Technology Zhejiang University of Technology Hangzhou310000 China College of Computer Science and Technology Zhejiang University of Technology Hangzhou310000 China

ISBN: (纸本)9781510681637

Lizards, ancient and fascinating reptiles, are active worldwide and play essential roles in ecosystems. However, their high level of similarity makes it challenging to identify species based on observation alone. Therefore, developing a simple and user-friendly set of criteria to aid in lizard species identification is an important and worthwhile research topic. In this study, we developed a universal discrimination model called SITS-Net for all lizard species and genders. It combines the Single Indicator Threshold Set (SITS) with decision trees. We initially divided the sample set randomly into subsets and used a greedy algorithm to maximize the purity of each subset. We then used a simulated annealing algorithm to optimize the threshold planning process. Finally, we integrated the planned threshold sets into a SITS-Net. We tested the model using a dataset and achieved a discrimination accuracy of over 95% for both genders of lizards and an overall species discrimination accuracy of 91.31%. The data-set used in the paper can be obtained at https://github. com/JiangYiL1EN/Lizards-Identification-Model. © 2024 SPIE.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Dynamic gesture recognition method based on multi-scale feature fusion 5

Dynamic gesture recognition method based on multi-scale feat...

引用

5th International Conference on computer Vision, Image and Deep Learning, CVIDL 2024

作者： Qiu, Feiyue Zhou, Jian Peng, Delong Zhejiang University of Technology College of Education Science and Technology Hangzhou310023 China Zhejiang University of Technology College of Computer Science and Technology Hangzhou310023 China

ISBN: (纸本)9798350373820

Currently, dynamic gesture classification methods based on deep learning have problems such as the inability to extract gesture features thoroughly and difficulty in accurate time series modeling of gesture features. This paper proposes a dynamic gesture recognition method based on multi-scale feature fusion to address these problems. First, dynamic gesture depth image sequences can be converted into 3D point cloud sequences to maintain the gesture's spatial structure and shape features. Then, a two-stream neural network structure is introduced to learn local and global features of gesture point cloud data separately, and they are fused into multi-scale features. To more accurately model dynamic gesture features in time series, a BiGRU-Transformer module is constructed by adding a bidirectional gated recurrent unit to the original structure of the Transformer, into which multi-scale features are input to obtain the final recognition results. To validate the effectiveness of our method, we perform experiments on SHREC17 and DHG datasets. The method achieves the highest recognition accuracy on these two datasets compared to other methods, validating the superiority of our proposed method. © 2024 IEEE.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Adaptive ROI Optimization Pyramid Network: Lane Detection for FSD under Data Uncertainty

引用

Engineering Letters 2025年第2期33卷 282-291页

作者： Cao, Xu Liu, Weisheng Wang, Zhijian College of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China College of Computer Science and Software Engineering University of Science and Technology Liaoning Co Anshan114051 China

To mitigate the challenges posed by data uncertainty in Full-Self Driving (FSD) systems. This paper proposes a novel feature extraction learning model called Adaptive Region of Interest Optimized Pyramid Network (ARO). Specifically, ARO introduces a novel cross-layer fusion attention mechanism that dynamically assigns weighted attention to feature maps across different levels, facilitating deep feature fusion and enabling the effective extraction and utilization of salient feature information. Furthermore, ARO incorporates a feature replication layer to duplicate and refine feature maps at multiple levels, thereby enhancing its capability to capture fine-grained details and generate richer feature representations. Additionally, a multi-path upsampling strategy preserves fine-grained features during upsampling. Extensive experimental evaluations conducted on the benchmark CuLane and Tusimple datasets demonstrate that ARO achieves an F1 score of 80.60% on the CuLane dataset, outperforming state-of-the-art methods, showcasing the effectiveness of the proposed approach in handling data uncertainty for robust lane detection in autonomous driving. Our code are available at https://***/caoxu0109/lane detect ldfr. © 2025, International Association of Engineers. All rights reserved.

关键词： Attention mechanisms Culane Lane detection Multi-scale fusion network Tusimple

来源：评论

学校读者我要写书评

暂无评论

Trademark detection and classification based on an improved YOLO

引用

Multimedia Tools and Applications 2025年 1-17页

作者： Miao, Chunyuan Wang, Xiuhui Department of Computer Science and Technology China Jiliang University Zhejiang Hangzhou310018 China

Trademarks are important identifiers for goods and services. They play an increasingly important role in daily life and production. However, with the continuous development of commercial society and the increasing number of goods and services, trademark detection and classification technology has become particularly critical, and it has large application prospects in the fields of commodity image search and trademark infringement. Based on the You Only Look Once (YOLO)-v5 framework, this paper proposes the YOLO-BC network to better detect trademark images. First, the network adopts a new bi-directional feature pyramid network & coordinate attention(BiFC) module to replace the original enhanced feature extraction network (PANet) model to improve the feature extraction ability of the network at different scales and capture key information. Second, a convolutional block attention module (CBAM) is introduced into the C3 module, and the influence of the C3-CBAM module embedded in the backbone and neck of the algorithm network on network performance is explored through experiments. Subsequently, the distance intersection over union non-maximum suppression (DIOU-NMS) is used to replace the NMS in the original network as the bounding box non-maximum suppression method. Finally, we replace generalized IOU (GIOU) with α-IOU as the bounding box regression loss function and use the appropriate parameter α to adaptively adjust the loss and gradient of the IOU target. Experimental results on public dataset and a self-constructed dataset show that the improved framework has a higher detection accuracy than traditional methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

A Time Series Short-Term Prediction Method Based on Multi-Granularity Event Matching and Alignment

引用

computers, Materials & Continua 2024年第1期78卷 653-676页

作者： Haibo Li Yongbo Yu Zhenbo Zhao Xiaokang Tang College of Computer Science and Technology Huaqiao UniversityXiamen361021China

Accurate forecasting of time series is crucial across various *** prediction tasks rely on effectively segmenting,matching,and time series data *** instance,regardless of time series with the same granularity,segmenting them into different granularity events can effectively mitigate the impact of varying time scales on prediction ***,these events of varying granularity frequently intersect with each other,which may possess unequal *** minor differences can result in significant errors when matching time series with future ***,directly using matched events but unaligned events as state vectors in machine learning-based prediction models can lead to insufficient prediction ***,this paper proposes a short-term forecasting method for time series based on a multi-granularity event,MGE-SP(multi-granularity event-based short-termprediction).First,amethodological framework for MGE-SP established guides the implementation *** framework consists of three key steps,including multi-granularity event matching based on the LTF(latest time first)strategy,multi-granularity event alignment using a piecewise aggregate approximation based on the compression ratio,and a short-term prediction model based on *** data from a nationwide online car-hailing service in China ensures the method’s *** average RMSE(root mean square error)and MAE(mean absolute error)of the proposed method are 3.204 and 2.360,lower than the respective values of 4.056 and 3.101 obtained using theARIMA(autoregressive integratedmoving average)method,as well as the values of 4.278 and 2.994 obtained using k-means-SVR(support vector regression)*** other experiment is conducted on stock data froma public data *** proposed method achieved an average RMSE and MAE of 0.836 and 0.696,lower than the respective values of 1.019 and 0.844 obtained using the ARIMA method,as well as the values of 1.350 and 1.172 obtained usin

关键词： Time series short-term prediction multi-granularity event alignment event matching

来源：评论

学校读者我要写书评

暂无评论

Finalizing your reference list with machine learning

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第11期14卷 14883-14892页

作者： Mei, Jian-Ping Chen, Defang Fan, Jing Yang, Lianghuai College of Computer Science and Technology Zhejiang University of Technology Hangzhou310023 China Jianxing Honors College Zhejiang University of Technology Hangzhou310023 China

A high quality reference list is important to the overall quality of a research paper. However, it requires domain knowledge and is time consuming to generate a reference list with good coverage, representativeness, and timeliness due to the large amount and fast growing of publications. In this paper, we deal with the specific problem of reference enhancement of research manuscripts with machine learning. A predictive model is trained by a large academic dataset with paper-related and venue-related information to discover additional references for a scientific draft with related information including an initial reference list. We propose a supervised approach called RefCom under the framework of learning-to-rank to predict the probability for a given paper to cite a reference candidate. Forty features in total are defined to describe pairs of papers with respect to author influence, venue influence and paper influence, as well as content and reference similarity. Unlike heuristic rule-based approaches, RefCom is able to integrate multiple features with learned weights. Experimental study with the AMiner dataset which contains 2 million papers and 1.7 million authors show the effectiveness of RefCom in citation prediction, suggesting its potential of being applied as an assistant tool in reference finalization. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Decomposition solutions and Bäcklund transformations of the B-type and C-type Kadomtsev-Petviashvili equations

引用

Communications in Theoretical Physics 2024年第6期76卷 34-44页

作者： Xiazhi Hao S Y Lou College of Science Zhejiang University of TechnologyHangzhou 310014China School of Physical Science and Technology Ningbo UniversityNingbo 315211China

This paper introduces a modified formal variable separation approach,showcasing a systematic and notably straightforward methodology for analyzing the B-type Kadomtsev-Petviashvili(BKP)*** the application of this approach,we successfully ascertain decomposition solutions,Bäcklund transformations,the Lax pair,and the linear superposition solution associated with the aforementioned ***,we expand the utilization of this technique to the C-type Kadomtsev-Petviashvili(CKP)equation,leading to the derivation of decomposition solutions,Bäcklund transformations,and the Lax pair specific to this *** results obtained not only underscore the efficacy of the proposed approach,but also highlight its potential in offering a profound comprehension of integrable behaviors in nonlinear ***,this approach demonstrates an efficient framework for establishing interrelations between diverse systems.

关键词： decomposition solution Backlund transformation Lax pair linear superposition solution

来源：评论

学校读者我要写书评

暂无评论

Improved Data Stream Clustering Method: Incorporating KD-Tree for Typicality and Eccentricity-Based Approach

引用

computers, Materials & Continua 2024年第2期78卷 2557-2573页

作者： Dayu Xu Jiaming Lu Xuyao Zhang Hongtao Zhang College of Mathematics and Computer Science Zhejiang A&F UniversityHangzhou311300China College of Economics and Management Zhejiang A&F UniversityHangzhou311300China

Data stream clustering is integral to contemporary big data ***,addressing the ongoing influx of data streams efficiently and accurately remains a primary challenge in current *** paper aims to elevate the efficiency and precision of data stream clustering,leveraging the TEDA(Typicality and Eccentricity Data Analysis)algorithm as a foundation,we introduce improvements by integrating a nearest neighbor search algorithm to enhance both the efficiency and accuracy of the *** original TEDA algorithm,grounded in the concept of“Typicality and Eccentricity Data Analytics”,represents an evolving and recursive method that requires no prior *** the algorithm autonomously creates and merges clusters as new data arrives,its efficiency is significantly hindered by the need to traverse all existing clusters upon the arrival of further *** work presents the NS-TEDA(Neighbor Search Based Typicality and Eccentricity Data Analysis)algorithm by incorporating a KD-Tree(K-Dimensional Tree)algorithm integrated with the Scapegoat *** arrival,this ensures that new data points interact solely with clusters in very close *** significantly enhances algorithm efficiency while preventing a single data point from joining too many clusters and mitigating the merging of clusters with high overlap to some *** apply the NS-TEDA algorithm to several well-known datasets,comparing its performance with other data stream clustering algorithms and the original TEDA *** results demonstrate that the proposed algorithm achieves higher accuracy,and its runtime exhibits almost linear dependence on the volume of data,making it more suitable for large-scale data stream analysis research.

关键词： Data stream clustering TEDA KD-tree scapegoat tree

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：