检索结果-内蒙古大学图书馆

3d model classification based on dRSN and multi-view feature fusion

EXPERT SYSTEMS WITH APPLICATIONS 2025年 273卷

作者： Gao, Xueyao Zhang, Yunkai Zhang, Chunxiang Xue, Yongzeng Harbin Univ Sci & Technol Sch Comp Sci & Technol Harbin 150080 Peoples R China Harbin Inst Technol Fac Comp Harbin 150001 Peoples R China

3d model classification based on view has become a hot research topic. If all projection views of 3d model are treated equally, the importance and difference of different views, and complementary and correlation information between views will be ignored. In order to solve these issues, this paper proposes a method of 3d model classification based on deep Residual Shrinkage Network (dRSN) and multi-view feature fusion. Firstly, 3d model is projected into six 2d views. Secondly, dRSN is used to extract view features from 2d views. Thirdly, shape distribution features d1, d2, d3 of 2d view are integrated with view features to get the fusion feature. Fourthly, the fusion feature is input into softmax function to get discriminative features and Shannon entropy is used to compute the uncertainty of view classification to measure view saliency. Fifthly, the fusion features of 2d views in descending order of view saliency are input into Long Short-Term Memory (LSTM) in sequence for fusing multi-view features. Finally, softmax function is adopted to classify 3d model based on multi-view fusion feature. Experimental results show that accuracy of the proposed method achieves 93.28% on modelNet10 dataset and it demonstrates higher accuracy.

关键词： 3d model classification Multi-view feature fusion Shape distribution features Shannon entropy View saliency

来源：评论

学校读者我要写书评

暂无评论

3d model classification based on regnet design space and voting algorithm

引用

MULTIMEdIA TOOLS ANd APPLICATIONS 2023年第14期83卷 42391页

作者： Gao, Xueyao Yan, Shaokang Zhang, Chunxiang Harbin Univ Sci & Technol Sch Comp Sci & Technol 52 Xuefu Rd Harbin 150080 Heilongjiang Peoples R China

3d models are widely used in industrial manufacturing, virtual reality, medical diagnosis and so on. At present, view-based 3d model classification has become an important research topic. However, single view feature can not describe the overall shape of 3d model. When multiple views are fused to describe 3d model, useful information is confused. It causes certain interference to determine 3d model's category. To solve these problems, a novel method of 3d model classification based on RegNet design space and voting algorithm is proposed. Firstly, 2d views of 3d model are input into RegNet design space with attention mechanism to extract high-level semantic feature(HSF). Secondly, HSF and the corresponding low-level shape features (LSF) of view are fused, including d1, d2, d3, Fourier descriptor, and Zernike moment. Thirdly, LSTM is combined with softmax function to extract more representative features from the fused feature. Finally, based on discriminative features, improved voting algorithm based on shannon entropy is constructed to determine 3d model's category. Experimental results show that average accuracy of the proposed method on modelNet10 reaches 94.93%, and the classification performance is outstanding.

关键词： 3d model classification Semantic feature RegNet design space Voting algorithm Shape features

来源：评论

学校读者我要写书评

暂无评论

Multi-Modal Meta-Transfer Fusion Network for Few-Shot 3d model classification

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2024年第3期132卷 673-688页

作者： Zhou, He-Yu Liu, An-An Zhang, Chen-Yu Zhu, Ping Zhang, Qian-Yi Kankanhalli, Mohan Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Natl Univ Singapore Sch Comp Singapore 117543 Singapore

Nowadays, driven by the increasing concern on 3d techniques, resulting in the large-scale 3d data, 3d model classification has attracted enormous attention from both research and industry communities. Most of the current methods highly depend on sufficient labeled 3d models, which substantially restricts their scalability to novel classes with few annotated training data since it can increase the chance of overfitting. Besides, they only leverage single-modal information (either point cloud or multi-view information), and few works integrate these complementary information for 3d model representation. To overcome these problems, we propose a multi-modal meta-transfer fusion network (M3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$<^>3d model classification$$\end{document}TF), the key of which is to perform few-shot multi-modal representation for 3d model classification. Specifically, we first convert the original 3d data into both multi-view and point cloud modalities, and pre-train individual encoding networks on a large-scale dataset to obtain the optimal initial parameters, which is beneficial to few-shot learning tasks. Then, to enable the network to adjust to few-shot learning tasks, we update the parameters in Scaling and Shifting operation (SS), multi-modal representation fusion (MMRF) and the 3d model classifier to obtain optimal initialization parameters. Since the large-scale training parameters in feature extractors will increase the chance of overfitting, we freeze the feature extractor and introduce a SS operation to adjust its weights. Specifically, SS can reduce the number of training parameters up to 20%, which can effectively avoid overfitting. MMRF can adaptively integrate the multi-modal information based on their significance to the 3d model for a more robust 3d representation. Since t

关键词： 3d model classification Multi-modal Multi-view Point cloud

来源：评论

学校读者我要写书评

暂无评论

Semantically guided projection for zero-shot 3d model classification and retrieval

引用

MULTIMEdIA SYSTEMS 2022年第6期28卷 2437-2451页

作者： Su, Yuting Li, Jiayu Li, Wenhui Gao, Zan Chen, Haipeng Li, Xuanya Liu, An-An Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230088 Peoples R China Qilu Univ Technol Shandong Artificial Intelligence Inst PR China Shandong Acad Sci Jinan Peoples R China Jilin Univ Sch Coll Comp Sci & Technol Jilin Jilin Peoples R China Baidu Inc Beijing Peoples R China

The most existing methods for 3d model classification and retrieval rely on the fully supervised training scheme, which are prohibitive and time-consuming to collect and label 3d models of wide different categories. How to make full use of the existing known data to represent the unknown data is a crucial topic. Inspired by the zero-shot learning in 2d image domain, we propose the semantically guided projection method to classify and retrieve unseen 3d models by exploring the semantic relationship between seen and unseen 3d models. First, we explore the multi-view information of 3d models to construct the semantic attributes as the prior knowledge to represent 3d models. Then, we learn bidirectional projections from visual features to semantics and from semantics to visual features, which can eliminate the gap between seen and unseen domains. Extensive experiments for zero-shot 3d model classification and retrieval on two popular datasets, modelNet40 and ShapeNetCore55, have demonstrated the effectiveness and superiority of the proposed method.

关键词： 3d model classification 3d model retrieval Multi-view zero-shot representation

来源：评论

学校读者我要写书评

暂无评论

3d model classification based on few-shot learning

引用

NEUROCOMPUTING 2020年 398卷 539-546页

作者： Nie, Jie Xu, Ning Zhou, Ming Yan, Ge Wei, Zhiqiang Ocean Univ China Coll Informat Sci & Engn Qingdao Peoples R China Tianjin Univ Sch Elect & Informat Engn Tianjin Peoples R China Chengdu Spaceon Measurement & Control Technol Co Chengdu Peoples R China

With the development of multimedia technology, 3d model has been applied in many fields such as mechanical design, construction industry, entertainment industry, medical treatment and so on. The number of 3d model is becoming more and more in our lives. Therefore, effective automatic management and classification of 3d models become more and more important. In this paper, we propose a dual-meta-learner model based on LSTM to learn the exact optimization algorithm used to train another two learner neural network classifier in the few-shot regime. The parametrization of our model allows it to learn appropriate parameter updates specifically for the scenario where a set amount of updates will be made, while it can also achieve a general initialization of the learner (classifier) network that allows for quick convergence of training. Our method attains state-of-the-art performance by significant margins. (C) 2019 Published by Elsevier B.V.

关键词： Few-shot Meta-learner 3d model classification

来源：评论

学校读者我要写书评

暂无评论

Multi-view SoftPool attention convolutional networks for 3d model classification

引用

FRONTIERS IN NEUROROBOTICS 2022年 16卷 1029968页

作者： Wang, Wenju Wang, Xiaolin Chen, Gang Zhou, Haoran Univ Shanghai Sci & Technol Coll Commun & Art Design Shanghai Peoples R China

IntroductionExisting multi-view-based 3d model classification methods have the problems of insufficient view refinement feature extraction and poor generalization ability of the network model, which makes it difficult to further improve the classification accuracy. To this end, this paper proposes a multi-view SoftPool attention convolutional network for 3d model classification tasks. MethodsThis method extracts multi-view features through ResNest and adaptive pooling modules, and the extracted features can better represent 3d models. Then, the results of the multi-view feature extraction processed using SoftPool are used as the Query for the self-attentive calculation, which enables the subsequent refinement extraction. We then input the attention scores calculated by Query and Key in the self-attention calculation into the mobile inverted bottleneck convolution, which effectively improves the generalization of the network model. Based on our proposed method, a compact 3d global descriptor is finally generated, achieving a high-accuracy 3d model classification performance. ResultsExperimental results showed that our method achieves 96.96% OA and 95.68% AA on modelNet40 and 98.57% OA and 98.42% AA on modelNet10. discussionCompared with a multitude of popular methods, our algorithm model achieves the state-of-the-art classification accuracy.

关键词： 3d model classification multi-view attention SoftPool convolutional

来源：评论

学校读者我要写书评

暂无评论

A novel neural network-based 3d animation model classification method

引用

INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS IN TECHNOLOGY 2023年第3期71卷 222-228页

作者： Shi, Ximan Xinxiang Univ Xinxiang Henan Peoples R China

The rapid development of information technology has also brought new vitality to art design. The 3d animation model making is a new multimedia technology based on computer technology. In order to efficiently organise and utilise the 3d model resources, researchers focus on how to achieve effective retrieval and classification. In order to realise the recognition and classification of 3d models, a novel network model called 3dSmallPCapsNet is proposed in this paper based on the feature that Capsule Network (CapsNet) exploits vector neurons to store feature space information. The proposed method can extract more representative features while reducing the model complexity. To evaluate our method, three different methods which are MeshNet, Shape-dNA and GPS-embedding, are compared. The experimental results on data sets SHREC10 and SHREC15 show that the proposed method has better performance.

关键词： 3d model classification CapsNet capsule network pooling animation model

来源：评论

学校读者我要写书评

暂无评论

Joint Local Correlation and Global Contextual Information for Unsupervised 3d model Retrieval and classification

引用

IEEE TRANSACTIONS ON CIRCUITS ANd SYSTEMS FOR VIdEO TECHNOLOGY 2022年第5期32卷 3265-3278页

作者： Li, Wenhui Zhao, Zhenlan Liu, An-An Gao, Zan Yan, Chenggang Mao, Zhendong Chen, Haipeng Nie, Weizhi Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Chinese Acad Sci Key Lab Electromagnet Space Informat Beijing 100864 Peoples R China Qilu Univ Technol Shandong Artificial Intelligence Inst Shandong Acad Sci Jinan 250353 Peoples R China Hangzhou Dianzi Univ Sch Automat Hangzhou 310018 Peoples R China Univ Sci & Technol China Sch Informat Sci & Technol Hefei 230052 Peoples R China Jilin Univ Coll Comp Sci & Technol Changchun 130012 Peoples R China

Unsupervised 3d model analysis has attracted tremendous attentions with the increasing growth of 3d model data and the extensive human annotations. Many effective methods have been designed to address the 3d model analysis with labeled information, while rare methods devote to unsupervised deep learning due to the difficulty of mining reliable information. In this paper, we propose a novel unsupervised deep learning method named joint local correlation and global contextual information (LCGC) for 3d model retrieval and classification, which mines the reliable triplet set and uses triplet loss to optimize the deep neural network. Our method proposes two schemes: 1) Local self-correlation information learning, which adopts the intra and inter information to construct the view-level triplet set. 2) Global neighbor contextual information learning, which employs the neighbor contextual information to explore the reliable relations among 3d models and construct the model-level triplet set. The above schemes encourage that the selected triple set can been used to improve the discrimination of learned features. Extensive evaluations on two large-scale datasets, modelNet40 and ShapeNet55, have demonstrated the effectiveness of our proposed method.

关键词： Three-dimensional displays Solid modeling Context modeling data models Analytical models deep learning Feature extraction 3d model retrieval 3d model classification unsupervised feature learning

来源：评论

学校读者我要写书评

暂无评论

Learning-Based Multiple Pooling Fusion in Multi-View Convolutional Neural Network for 3d model classification and Retrieval

引用

JOURNAL OF INFORMATION PROCESSING SYSTEMS 2019年第5期15卷 1179-1191页

作者： Zeng, Hui Wang, Qi Li, Chen Song, Wei Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing Engn Res Ctr Ind Spectrum Imaging Beijing Peoples R China North China Univ Technol Sch Comp Sci & Technol Beijing Peoples R China

We design an ingenious view-pooling method named learning-based multiple pooling fusion (LMPF), and apply it to multi-view convolutional neural network (MVCNN) for 3d model classification or retrieval. By this means, multi-view feature maps projected from a 3d model can be compiled as a simple and effective feature descriptor. The LMPF method fuses the max pooling method and the mean pooling method by learning a set of optimal weights. Compared with the hand-crafted approaches such as max pooling and mean pooling, the LMPF method can decrease the information loss effectively because of its "learning" ability. Experiments on modelNet40 dataset and McGill dataset are presented and the results verify that LMPF can outperform those previous methods to a great extent.

关键词： Learning-Based Multiple Pooling Fusion Multi-View Convolutional Neural Network 3d model classification 3d model Retrieval

来源：评论

学校读者我要写书评

暂无评论

Category-specific upright orientation estimation for 3d model classification and retrieval

引用

IMAGE ANd VISION COMPUTING 2020年 96卷 103900-000页

作者： Kim, Seong-heum Hwang, Youngbae Kweon, In So Korea Elect Technol Inst Intelligent Image Proc Res Ctr Seongnam Si 13509 Gyeonggi Do South Korea Chungbuk Natl Univ Dept Elect Engn 1 Chungdae Ro Cheongju 28644 Chungbuk South Korea Korea Adv Inst Sci & Technol Elect Engn Dept 335 Gwahak Ro Daejeon 305701 South Korea

In this paper, we address a problem of correcting upright orientation of a reconstructed object to search. We first reconstruct an input object appearing in an image sequence, and generate a query shape using multi-view object co-segmentation. In the next phase, we utilize the Convolutional Neural Network (CNN) architecture to determine category-specific upright orientation of the queried shape for 3d model classification and retrieval. As a practical application of our system, a shape style and a pose from an inferred category and up-vector are obtained by comparing 3d shape similarity with candidate 3d models and aligning its projections with a set of 2d co-segmentation masks. We quantitatively and qualitatively evaluate the presented system with more than 720 upfront-aligned 3d models and five sets of multi-view image sequences. (C) 2020 Published by Elsevier B.V.

关键词： model-based 3d reconstruction Multi-view object co-segmentation Convolutional neural networks Upright orientation estimation 3d model classification 3d model classification retrieval

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

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

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：