检索结果-内蒙古大学图书馆

VLdB JOURNAL 2025年第1期34卷 1-27页

作者： Liu, Hao Wong, Raymond Chi-Wing Hong Kong Univ Sci & Technol Dept Comp Sci & Engn Hong Kong Peoples R China

due to the growth of the 3d technology, digital 3d models represented in the form of point clouds have attracted a lot of attention from both industry and academia. In this paper, due to a variety of applications, we study a fundamental problem called the 3d object retrieval, which is to find a set of 3d point clouds stored in a database that are similar to a given query 3d point cloud. To the best of our knowledge, solving the problem of 3d object retrieval efficiently remains unexplored in the research community. In this paper, we propose a framework called C2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document}O to find the answer efficiently with the help of an index built on the database. In most of our experiments, C2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document}O performs up to 2 orders of magnitude faster than all adapted algorithms in the literature. In particular, when the database size scales up to 100 million points, C2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document}O answers the 3d object retrieval within 10 s but all adapted exact algorithms need more than 1000 s.

关键词： Point cloud 3d object retrieval donut representation rotation translation invariance database

来源：评论

学校读者我要写书评

暂无评论

Hypergraph-Based Multi-Modal Representation for Open-Set 3d object retrieval

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS ANd MACHINE INTELLIGENCE 2024年第4期46卷 2206-2223页

作者： Feng, Yifan Ji, Shuyi Liu, Yu-Shen du, Shaoyi dai, Qionghai Gao, Yue Tsinghua Univ Sch Software BN Rist THUIBCSBLBCI Beijing 100084 Peoples R China Tsinghua Univ Sch Software Beijing 100084 Peoples R China Tsinghua Univ BNRist THUIBCS BLBCI Beijing 100084 Peoples R China Xi An Jiao Tong Univ Coll Artificial Intelligence Inst Artificial Intelligence & Robot Xian 710049 Peoples R China

The traditional 3d object retrieval (3dOR) task is under the close-set setting, which assumes the categories of objects in the retrieval stage are all seen in the training stage. Existing methods under this setting may tend to only lazily discriminate their categories, while not learning a generalized 3d object embedding. Under such circumstances, it is still a challenging and open problem in real-world applications due to the existence of various unseen categories. In this paper, we first introduce the open-set 3dOR task to expand the applications of the traditional 3dOR task. Then, we propose the Hypergraph-Based Multi-Modal Representation (HGM 22 R) framework to learn 3d object embeddings from multi-modal representations under the open-set setting. The proposed framework is composed of two modules, i.e., the Multi-Modal 3d object Embedding (MM3dOE) module and the Structure-Aware and Invariant Knowledge Learning (SAIKL) module. By utilizing the collaborative information of modalities derived from the same 3d object, the MM3dOE module is able to overcome the distinction across different modality representations and generate unified 3d object embeddings. Then, the SAIKL module utilizes the constructed hypergraph structure to model the high-order correlation among 3d objects from both seen and unseen categories. The SAIKL module also includes a memory bank that stores typical representations of 3d objects. By aligning with those memory anchors in the memory bank, the aligned embeddings can integrate the invariant knowledge to exhibit a powerful generalized capacity toward unseen categories. We formally prove that hypergraph modeling has better representative capability on data correlation than graph modeling. We generate four multi-modal datasets for the open-set 3dOR task, i.e., OS-ESB-core, OS-NTU-core, OS-MN40-core, and OS-ABO-core, in which each 3d object contains three modality representations: multi-view, point clouds, and voxel. Experiments on these four datas

关键词： Three-dimensional displays Task analysis Solid modeling Point cloud compression Correlation data models Feature extraction Hypergraph multi-modal 3d object retrieval open-set memory bank

来源：评论

学校读者我要写书评

暂无评论

Multi-Range View Aggregation Network With Vision Transformer Feature Fusion for 3d object retrieval

引用

IEEE TRANSACTIONS ON MULTIMEdIA 2023年 25卷 9108-9119页

作者： Lin, dongyun Li, Yiqun Cheng, Yi Prasad, Shitala Guo, Aiyuan Cao, Yanpeng ASTAR Inst Infocomm Res I2R Singapore 138632 Singapore Zhejiang Univ State Key Lab Fluid Power & Mechatron Syst Hangzhou 310027 Peoples R China

View-based methods have achieved state-of-the-art performance in 3d object retrieval. However, view-based methods still encounter two major challenges. The first is how to leverage the inter-view correlation to enhance view-level visual features. The second is how to effectively fuse view-level features into a discriminative global descriptor. Towards these two challenges, we propose a multi-range view aggregation network (MRVA-Net) with a vision transformer based feature fusion scheme for 3d object retrieval. Unlike the existing methods which only consider aggregating neighboring or adjacent views which could bring in redundant information, we propose a multi-range view aggregation module to enhance individual view representations through view aggregation beyond only neighboring views but also incorporate the views at different ranges. Furthermore, to generate the global descriptor from view-level features, we propose to employ the multi-head self-attention mechanism introduced by vision transformer to fuse the view-level features. Extensive experiments conducted on three public datasets including ModelNet40, ShapeNet Core55 and MCB-A demonstrate the superiority of the proposed network over the state-of-the-art methods in 3d object retrieval.

关键词： Three-dimensional displays Feature extraction Transformers Convolutional neural networks Visualization Fuses deep learning 3d object retrieval multi-range view aggregation multi-head self-attention feature fusion

来源：评论

学校读者我要写书评

暂无评论

Vulnerability of Feature Extractors in 2d Image-Based 3d object retrieval

引用

IEEE TRANSACTIONS ON MULTIMEdIA 2023年 25卷 5065-5076页

作者： Liu, An-An Zhou, He-Yu Li, Xuanya Wang, Lanjun Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230088 Peoples R China Baidu Inc Beijing 100105 Peoples R China Tianjin Univ Sch New Media & Commun Tianjin 300072 Peoples R China

Recent advances in 3d modeling software and 3d capture devices contribute to the availability of large-scale 3d objects. Together with the prevalence of deep neural networks (dNNs), dNN-based 3d object retrieval systems are widely applied, especially by inputting 2d images to retrieve 3d objects. Although dNNs have shown vulnerable to adversarial attacks in classification, the vulnerability of dNN-based 3d object retrieval system remains under-explored. In this paper, we formulate the problem of attacking against dNN-based feature extractors in the 2d image-based 3d object retrieval system. Specifically, we consider the attack happens under a reasonable scenario that the candidate 3d object database is unknown to the adversary, which challenges adversarial example generation. To tackle this difficulty, we set up a reasonable hypothesis on the information which the adversary can be accessible, and then propose two effective perturbation generation methods: one is to corrupt domain-level alignment (CdA) and the other one is to corrupt class-level alignment (CCA). In converse, we propose a novel progressive adversarial training (PAT) method to improve the feature extractor robustness, which can effectively and stably mitigate both CdA and CCA attacks. Experimental results demonstrate that a typical feature extractor can be effectively compromised by attacks. Moreover, the transferability of the adversarial query illustrates the possibility of realistic black-box attacks. The successful defense against both CdA and CCA attacks by PAT can validate the superiority of the proposed defense method.

关键词： 3d object retrieval adversarial example adversarial defense

来源：评论

学校读者我要写书评

暂无评论

SHREC'22 track: Open-Set 3d object retrieval

引用

COMPUTERS & GRAPHICS-UK 2022年 107卷 231-240页

作者： Feng, Yifan Gao, Yue Zhao, Xibin Guo, Yandong Bagewadi, Nihar Bui, Nhat-Tan dao, Hieu Gangisetty, Shankar Guan, Ripeng Han, Xie Hua, Cong Hunakunti, Chidambar Jiang, Yu Jiao, Shichao Ke, Yuqi Kuang, Liqun Liu, Anan Nguyen, dinh-Huan Nguyen, Hai-dang Nie, Weizhi Pham, Bang-dang Raikar, Karthik Tang, Qingmei Tran, Minh-Triet Wan, Jialong Yan, Chenggang You, Haoxuan Zhu, difei Tsinghua Univ Sch Software Beijing Peoples R China Tsinghua Univ BNRist THUIBCS KLISSBLBCI Beijing Peoples R China OPPO Res Inst Beijing Peoples R China Univ Sci Ho Chi Minh City Vietnam John Neumann Inst Ho Chi Minh City Vietnam Vietnam Natl Univ Ho Chi Minh City Vietnam KLE Technol Univ Hubballi India North Univ China Sch Data Sci & Technol Taiyuan Peoples R China Tianjin Univ Sch Elect & Informat Engn Tianjin Peoples R China Zhejiang Univ Sch Management Hangzhou Zhejiang Peoples R China Shandong Univ Sch Mech Elect Informat Engn Jinan Peoples R China Jilin Univ Coll Comp Sci & Technol Changchun Peoples R China Columbia Univ New York NY 10025 USA

This paper reports the results of the SHREC'22 track: Open-Set 3d object retrieval, the goal of which is to evaluate the performance of different retrieval algorithms under the Open-Set setting and modality-missing setting, respectively. Since objects from unseen categories are very common in real-world applications, we design the open-set 3d object retrieval to expand the application of traditional 3d object retrieval. In this track, we generate open-set 3d object retrieval datasets OS-MN40 and OS-MN40-Miss based on the ModelNet40 dataset, which are collected for the open-set setting and both open-set setting and modality-missing setting, respectively. Both the two datasets include the training set (2822 objects from 8 categories) and the retrieval set (960 query objects and 8527 target objects from the other 32 categories). The categories of retrieval (query/target) sets are not seen in the training set. For each object in the OS-MN40, four types of modalities, including mesh, point cloud, multi-view, and voxel, are provided. Each object in the OS-MN40-Miss is represented with incomplete modality information, which is collected to simulate the retrieval task in the real world. This track attracted eight participants from four countries and 191 runs of all submissions. The evaluation results show a promising scenario about open-set retrieval on 3d objects with multi -modal and multi-resolution representation and reveal interesting insights in dealing with retrieving 3d objects in unknown-category objects.(c) 2022 Elsevier Ltd. All rights reserved.

关键词： SHREC?22 3d object retrieval Open-Set Multi-Modal

来源：评论

学校读者我要写书评

暂无评论

Feature representation for 3d object retrieval based on unconstrained multi-view

引用

MULTIMEdIA SYSTEMS 2022年第5期28卷 1699-1711页

作者： Zhou, Bin Wang, Xuanyin Zhejiang Univ State Key Lab Fluid Power & Mechatron Syst 38 Zheda Rd Hangzhou 310027 Zhejiang Peoples R China

Reasonable and accurate image feature representation is the key to successful object retrieval. In this paper, we propose a 3d object feature representation method based on multiple views rather than a shape model. Unlike existing view-based methods that use pre-designed camera arrays to capture views, our method is flexible to implement by using several unconstrained views. Firstly, we generate a histogram of word frequencies to represent each view through local feature quantization. Then we integrate the histogram vectors of views belonging to the same object to generate a complete feature representation. Finally, similarity between two features is calculated for object retrieval. Several criteria are employed to evaluate the retrieval quality of the proposed method. Experimental results show that the integrated model feature is more effective and efficient than a set of individual image features and our approach is also competitive among several state-of-the-art methods.

关键词： 3d object retrieval Multi-view Feature representation Bag-of-words Histogram vector

来源：评论

学校读者我要写书评

暂无评论

Triadic Elastic Structure Representation for Open-Set Incremental 3d object retrieval 24

Triadic Elastic Structure Representation for Open-Set Increm...

引用

4th Annual International Conference on Multimedia retrieval (ICMR)

作者： Xu, Yang Feng, Yifan Bie, Lin Tsinghua Univ Sch Software Beijing Peoples R China

ISBN: (纸本)9798400706028

In open-set environments, the introduction of unseen classes adds a third stage of data for incremental learning, alongside seen classes of old and new. Existing methods by binary optimization between old and new classes struggle to model the complex triadic relationships among old, new, and unseen classes. In this paper, we introduce the Triadic Elastic Structure Learning (TESR) framework for open-set incremental 3d object retrieval. Specifically, to overcome the Bi-directional Catastrophic Forgetting (BdCF) problem in open-set incremental learning, we employ the Triadic Regularized Embedding (TRE) module. This module helps to globally preserve the triadic relationships among the three stages of classes, while simultaneously facilitating semantic-specific learning for new classes from a local perspective. To address the challenge of Bi-directional Knowledge Transfer (BdKT), our method leverages high-order correlations among objects of different classes through the Elastic Structure distillation (ESd) module, by constructing an elastic hypergraph based on incremental and global correlations. We construct four multi-modal datasets for this open-set incremental 3d object retrieval: OIES, OINT, OIMN, and OIAB. Extensive experiments and ablation studies on these four benchmarks demonstrate the superiority of our method over current state-of-the-art approaches.

关键词： 3d object retrieval Open-Set Learning Incremental Learning Multi-Modal Fusion Hypergraph

来源：评论

学校读者我要写书评

暂无评论

2dSlicesNet: A 2d Slice-Based Convolutional Neural Network for 3d object retrieval and Classification

引用

IEEE ACCESS 2021年 9卷 24041-24049页

作者： Taybi, Ilyass Ouazzani Gadi, Taoufiq Alaoui, Rachid Hassan First Univ Fac Sci & Tech LIIMSC Lab Settat 26000 Morocco Mohammed V Univ Fac Sci LRIT Lab Rabat 10000 Morocco Mohammed V Univ Higher Sch Technol Sale LASTIMI Lab Rabat 10000 Morocco

3d data can be instrumental to the computer vision field as it provides insightful information about the full 3d models' geometry. Recently, with easy access to both computational power and huge 3d databases, it is feasible to apply convolutional neural networks to automatically extract the 3d models' features. This paper presents a novel approach, called 2dSlicesNet, which deals with the issue of 3d model retrieval and classification using a 2d slice-based representation with a 3d convolutional neural network. The assumption in this context is that similar 3d models will be composed of almost identical 2d slices. Therefore, we first transform each normalized 3d model into a set of 2d slices corresponding to its first main axis, and then use them as input data to our 3d convolutional neural network. Experimental results and comparison with state-of-the-art approaches, using ModelNet10 and ModelNet40 datasets, prove that our proposed 2dSlicesNet approach can reach notable rates of accuracy in classification and retrieval.

关键词： Three-dimensional displays Two dimensional displays Solid modeling Convolution Computational modeling Feature extraction Convolutional neural networks deep learning 2d slices 3d convolutional neural network 3d object classification 3d object retrieval

来源：评论

学校读者我要写书评

暂无评论

3d object retrieval based on non-local graph neural networks

引用

MULTIMEdIA TOOLS ANd APPLICATIONS 2020年第45-46期79卷 34011-34027页

作者： Li, Yin-min Gao, Zan Tao, Ya-bin Wang, Li-li Xue, Yan-bing Tianjin Univ Technol Tianjin Key Lab Intelligence Comp & Novel Softwar Key Lab Comp Vis & Syst Minist Educ Tianjin 300384 Peoples R China Qilu Univ Technol Shandong Artif Intelligence Inst Shandong Acad Sci Jinan 250014 Peoples R China Jiangxi Vocat Tech Coll Ind Trade Nanchang 330038 Jiangxi Peoples R China China Unicorn Yantai Branch Yantai 264006 Peoples R China

3d object retrieval is a hot research field in computer vision and multimedia analysis domain. Since the appearance feature and points of view of 3d objects are very different, thus, the distribution of the training set and test set are variant which is very suitable for transfer learning or cross-domain learning. In the transfer learning or cross-domain learning, the feature extraction is very important which should have good robust for different domains. Thus, in this work, we pay attention to the feature extraction of 3d objects. So far, different feature representations and object retrieval approaches have been proposed. Among them, view-based deep learning retrieval methods achieve state-of-the-art performance, but the existing deep learning retrieval methods only simply use a deep neural network to extract features from each view and directly obtain the view-level shape descriptors without utilizing the spatial relationship between the views. In order to mine the spatial relationship among different views and obtain more discriminative 3d shape descriptors, in this work, 3d object retrieval based on non-local graph neural networks (NGNN) is proposed. In detail, the residual network is firstly utilized as the infrastructure, and then the non-local structure is embedded in the resnet to learn the intrinsic relationship between the views. Finally, the view pooling layer is employed to further fuse the information from different views, and obtain the discriminate feature for the 3d object. Experimental results on two public MVREd and NTU 3d datasets show that the non-local graph network is very efficient for exploring the latent relationship among different views, and the performance ofNGNNsignificantly outperforms state-of-the-art approaches whose improvement can reaches 12.4%-22.7% on ANMRR.

关键词： 3d object retrieval Non-local graph neural network 3d shape descriptors

来源：评论

学校读者我要写书评

暂无评论

Balanced Class-Incremental 3d object Classification and retrieval

引用

IEEE TRANSACTIONS ON KNOWLEdGE ANd dATA ENGINEERING 2024年第1期36卷 35-48页

作者： Liu, An-An Lu, Haochun Zhou, Heyu Li, Tianbao Kankanhalli, Mohan Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230088 Anhui Peoples R China Natl Univ Singapore Sch Comp Singapore 117543 Singapore

Most existing 3d object classification and retrieval algorithms rely on one-off supervised learning on closed 3d object sets and tend to provide rigid convolutional neural networks with little scalability. Such limitations substantially restrict their potential to learn newly emerged 3d object classes continually in the real world. Aiming to go beyond these limitations, we innovatively propose two new and challenging tasks: class-incremental 3d object classification (CI-3dOC) and class-incremental 3d object retrieval (CI-3dOR), the key to which is class-incremental 3d representation learning. It expects the network to update continually to learn new 3d class representations without forgetting the previously learned ones. To this end, we design a novel balanced distillation network (BdNet) that uses a dual supervision mechanism to balance between consolidating old knowledge (stability) and adapting to new 3d object classes (plasticity) carefully. On the one hand, we employ stability-based supervision to retain the stable and discriminative information of old classes that greatly benefit both classification and retrieval tasks. On the other hand, we use plasticity-based supervision to improve the network's generalization for learning new class 3d representations by transferring knowledge from a temporary teacher network to the current model. By properly handling the relationship between the two modules, we achieve a surprising performance improvement. Furthermore, considering there is no available dataset for evaluation, we build two 3d datasets, INOR-1 and INOR-2, to evaluate these two new tasks. Extensive experimental results demonstrate that our method can significantly outperform other state-of-the-art class-incremental learning methods. Even if we store 500-1000 fewer 3d objects than SOTA methods, BdNet still achieves comparable performance.

关键词： 3d representation learning 3d object classification 3d object retrieval class-incremental learning

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：