检索结果-内蒙古大学图书馆

Multi-Head Structural Attention-Based Vision Transformer with Sequential Views for 3d object recognition

APPLIEd SCIENCES-BASEL 2025年第6期15卷 3230-3230页

作者： Bao, Jianjun Luo, Ke Kou, Qiqi He, Liang Zhao, Guo Tiandi Changzhou Automat Co Ltd Changzhou 213015 Peoples R China China Univ Min & Technol Sch Comp Sci & Technol Xuzhou 221116 Peoples R China Beihang Univ Sch Artificial Intelligence Beijing 100191 Peoples R China Beijing HuaHang Inst Radio Measurement Beijing 100013 Peoples R China

Multi-view image classification tasks require the effective extraction of both spatial and temporal features to fully leverage the complementary information across views. In this study, we propose a lightweight yet powerful model, Multi-head Sparse Structural Attention-based Vision Transformer (MSSAViT), which integrates Structural Self-Attention mechanisms into a compact framework optimized for multi-view inputs. The model employs a fixed MobileNetV3 as a Feature Extraction Module (FEM) to ensure consistent feature patterns across views, followed by Spatial Sparse Self-Attention (SSSA) and Temporal Sparse Self-Attention (TSSA) modules that capture long-range spatial dependencies and inter-view temporal dynamics, respectively. By leveraging these structural attention mechanisms, the model achieves the effective fusion of spatial and temporal information. Importantly, the total model size is reduced to 6.1 M with only 1.5 M trainable parameters, making it highly efficient. Comprehensive experiments demonstrate the proposed model's superior performance and robustness in multi-view classification tasks, outperforming baseline methods while maintaining a lightweight design. These results highlight the potential of MSSAViT as a practical solution for real-world applications under resource constraints.

关键词： 3d object recognition sequential views self-attention mechanism vision transformer

来源：评论

学校读者我要写书评

暂无评论

CFMVOR: Federated Multi-view 3d object recognition Based on Compressed Learning 7th

CFMVOR: Federated Multi-view 3D Object Recognition Based on ...

引用

7th Chinese Conference on Pattern recognition and Computer Vision

作者： Xiao, di Zhang, Meng Zhang, Maolan Chen, Lvjun Chongqing Univ Chongqing 400044 Peoples R China

ISBN: (纸本)9789819784929;9789819784936

In distributed scenarios, the implementation of multi-view 3d object recognition algorithms through federated analytics (FA) frequently results in communication costs induced by feature aggregation that significantly contribute to the overall latency and energy consumption of the system. Furthermore, the transmission of features may give rise to concerns regarding privacy. This paper presents a federated multi-view 3d object recognition scheme based on compressed learning (called as CFMVOR). Firstly, in order to reduce the communication overhead, compressed learning is applied in order to extract features directly from compressed sensing measurements. Concurrently, the measurements appear visually similar to noise, thereby providing visual privacy. Subsequently, we present the hierarchical feature aggregation (HFA) method, which reduces the volume of communication features without compromising model accuracy. Finally, to address the observed threats to FA, we introduce local differential privacy (LdP) and enhance its perturbation mechanism to achieve feature-level adaptive LdP protection. Experiments on the ModelNet40 and ModelNet10 datasets demonstrate that our scheme, with a minimum compression ratio of 0.1, can significantly reduce communication overhead and, with a lower privacy budget epsilon = 5, maintain accuracy above 90% on both datasets.

关键词： 3d object recognition view-based methods compressed learning federated analytics local differential privacy

来源：评论

学校读者我要写书评

暂无评论

3d object recognition and Pose Estimation From Point Cloud Using Stably Observed Point Pair Feature

引用

IEEE ACCESS 2020年 8卷 44335-44345页

作者： Li, deping Wang, Hanyun Liu, Ning Wang, Xiaoming Xu, Jin Jinan Univ Coll Informat Sci & Technol Guangzhou 510632 Guangdong Peoples R China Jinan Univ Robot Res Inst Guangzhou 510632 Guangdong Peoples R China Informat Engn Univ Inst Surveying & Mapping Zhengzhou 450000 Henan Peoples R China

recognition and pose estimation from 3d free-form objects is a key step for autonomous robotic manipulation. Recently, the point pair features (PPF) voting approach has been shown to be effective for simultaneous object recognition and pose estimation. However, the global model descriptor (e.g., PPF and its variants) that contained some unnecessary point pair features decreases the recognition performance and increases computational efficiency. To address this issue, in this paper, we introduce a novel strategy for building a global model descriptor using stably observed point pairs. The stably observed point pairs are calculated from the partial view point clouds which are rendered by the virtual camera from various viewpoints. The global model descriptor is extracted from the stably observed point pairs and then stored in a hash table. Experiments on several datasets show that our proposed method reduces redundant point pair features and achieves better compromise of speed vs accuracy.

关键词： Three-dimensional displays Feature extraction Solid modeling object recognition Pose estimation Observability Computational modeling 3d object recognition 3d pose estimation point cloud point pair feature

来源：评论

学校读者我要写书评

暂无评论

3d object recognition method with multiple feature extraction from LidAR point clouds

引用

JOURNAL OF SUPERCOMPUTING 2019年第8期75卷 4430-4442页

作者： Tian, Yifei Song, Wei Sun, Su Fong, Simon Zou, Shuanghui North China Univ Technol 5 Jinyuanzhuang Rd Beijing 100144 Peoples R China Univ Macau Dept Comp & Informat Sci Taipa 999078 Macau Peoples R China Beijing Key Lab Urban Intelligent Traff Control T Beijing 100144 Peoples R China

during autonomous driving, fast and accurate object recognition supports environment perception for local path planning of unmanned ground vehicles. Feature extraction and object recognition from large-scale 3d point clouds incur massive computational and time costs. To implement fast environment perception, this paper proposes a 3d recognition system with multiple feature extraction from light detection and ranging point clouds modified by parallel computing. Effective object feature extraction is a necessary step prior to executing an object recognition procedure. In the proposed system, multiple geometry features of a point cloud that resides in corresponding voxels are computed concurrently. In addition, a scale filter is employed to convert feature vectors from uncertain count voxels to a normalized object feature matrix, which is convenient for object-recognizing classifiers. After generating the object feature matrices of all voxels, an initialized multilayer neural network (NN) model is trained offline through a large number of iterations. Using the trained NN model, real-time object recognition is realized using parallel computing technology to accelerate computation.

关键词： 3d object recognition Feature extraction LidAR point cloud Parallel computing

来源：评论

学校读者我要写书评

暂无评论

3d object recognition using deep learning for automatically generating semantic BIM data

引用

AUTOMATION IN CONSTRUCTION 2024年 162卷

作者： Rogage, Kay doukari, Omar Northumbria Univ Fac Engn & Environm Newcastle Upon Tyne NE1 8ST England

The successful reuse of Building Information Model (BIM) data is reliant on the use of clearly defined objects. File formats such as the Industry Foundation Classes (IFC) along with classification systems offer approaches to standardising semantic BIM data. Inconsistent application of these standards during BIM authoring results in objects that are unable to be reused across systems. This research proposes a Machine Learning (ML) based alternative for detecting objects and enriching BIM data based on geometrical representations of objects to support interoperability across systems throughout the supply chain. First, an existing dataset of training and test data is modified to correspond with predefined data types from the IFC schema. Second, a convolutional deep belief network is implemented and applied on new testing data from the National BIM Library to identify 3d objects and enabling semantic enrichment of BIM data. Finally, the testing is extended to four real-world industry BIM models. The findings show that the geometric data from emerging BIM object libraries can be used to search and retrieve objects within industry BIM models based on geometric rather than semantic data.

关键词： BIM Standardisation Information design quality Asset management 3d object recognition deep learning

来源：评论

学校读者我要写书评

暂无评论

3d object recognition based on pairwise Multi-view Convolutional Neural Networks

引用

JOURNAL OF VISUAL COMMUNICATION ANd IMAGE REPRESENTATION 2018年 56卷 305-315页

作者： Gao, Z. Wang, d. Y. Xue, Y. B. Xu, G. P. Zhang, H. Wang, Y. L. Qilu Univ Technol Shandong Acad Sci Shandong Artificial Intelligence Inst Jinan 250014 Shandong Peoples R China Natl Supercomp Ctr Jinan Shandong Comp Sci Ctr Jinan 250014 Shandong Peoples R China Tianjin Univ Technol Minist Educ Key Lab Comp Vis & Syst Tianjin 300384 Peoples R China Tianjin Univ Technol Tianjin Key Lab Intelligence Comp & Novel Softwar Tianjin 300384 Peoples R China

With the development of 3d sensors, it will be much easier for us to obtain 3d models, which is prevailing in our future daily life, but up to now, although many 3d object recognition algorithms have been proposed, there are some limitations, including the lack of training samples, hand-crafted feature representation, feature extraction and recognition separately. In this work, we propose a novel pairwise Multi-View Convolutional Neural Network for 3d object recognition (PMV-CNN for short), where automatic feature extraction and object recognition are put into a unify CNN architecture. Moreover, since the pairwise network architecture is utilized in PMV-CNN, thus, the requirement of the number of training samples in the original dataset is not severe. In addition, the latent complementary relationships from different views can be highly explored by view pooling. Large scale experiments demonstrate that the pairwise architecture is very useful when the number of labeled training samples is very small. Moreover, it also makes more robust feature extraction. Furthermore, since the end-to-end network architecture is employed in PMV-CNN, thus, the extracted feature is very suitable for 3d object recognition, whose performance is much better than that of hand-crafted features. In a word, the performance of our proposed method outperforms state-of-the-art methods. (C) 2018 Elsevier Inc. All rights reserved.

关键词： 3d object recognition CNN Multi-view PMV-CNN Pairwise End-to-end

来源：评论

学校读者我要写书评

暂无评论

3d object recognition using scale-invariant features

引用

VISUAL COMPUTER 2019年第1期35卷 71-84页

作者： Lim, Jeonghun Lee, Kunwoo Seoul Natl Univ Inst Engn Res Sch Mech Engn Seoul South Korea

As 3d scanning technology develops, it becomes easier to acquire various 3d surface data;thus, there is a growing need for 3d data registration and recognition technology. Many existing studies use local descriptors using local surface patches, and most of them use a fixed support radius, so they cannot cope perfectly when the model and scene have different scales. In this study, we propose a perfectly scale-invariant feature selection algorithm by extending the 2d SIFT algorithm (Lowe in Int J Comput Vis 60(2):91-110, 2004) to a 3d mesh. The feature selection method proposed in this study can obtain highly repeatable feature points and support radii regardless of mesh scale. The selected features can effectively describe the local information by the new shape descriptor proposed in this study. Unlike existing shape descriptors, it is possible to perform scale-invariant 3d object recognition and achieve a high recognition rate when combined with the feature point selection algorithm proposed in this study by using the gradients of the scalar functions defined on the 3d surface. We also reduced the searching space and lowered the false positive rate by suggesting a new RANSAC-based transformation hypotheses generation algorithm. Our 3d object recognition algorithm achieves recognition rates of 100 and 98.5%, respectively, when tested on the U3OR and CFVd datasets, exceeding the results of previous studies.

关键词： 3d feature selection 3d feature descriptor 3d object recognition Scale-invariant recognition

来源：评论

学校读者我要写书评

暂无评论

3d object recognition in cluttered environments by segment-based stereo vision

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2002年第1期46卷 5-23页

作者： Sumi, Y Kawai, Y Yoshimi, T Tomita, F Natl Inst Adv Ind Sci & Technol Intelligent Syst Inst AIST Tsukuba Cent 2 Tsukuba Ibaraki 3058568 Japan

We propose a new method for 3d object recognition which uses segment-based stereo vision. An object is identified in a cluttered environment and its position and orientation (6 dof) are determined accurately enabling a robot to pick up the object and manipulate it. The object can be of any shape (planar figures, polyhedra, free-form objects) and partially occluded by other objects. Segment-based stereo vision is employed for 3d sensing. Both CAd-based and sensor-based object modeling subsystems are available. Matching is performed by calculating candidates for the object position and orientation using local features, verifying each candidate, and improving the accuracy of the position and orientation by an iteration method. Several experimental results are presented to demonstrate the usefulness of the proposed method.

关键词： 3d object recognition free-form objects segment-based stereo vision 3d object modeling 3d shape matching robot vision

来源：评论

学校读者我要写书评

暂无评论

3d object recognition and classification: a systematic literature review

引用

PATTERN ANALYSIS ANd APPLICATIONS 2019年第4期22卷 1243-1292页

作者： Carvalho, L. E. von Wangenheim, A. Univ Fed Santa Catarina Grad Program Comp Sci Florianopolis SC Brazil Univ Fed Santa Catarina Natl Brazilian Inst Digital Convergence Image Proc & Comp Graph Lab Florianopolis SC Brazil

In this paper, we present a systematic literature review concerning 3d object recognition and classification. We cover articles published between 2006 and 2016 available in three scientific databases (Sciencedirect, IEEE Xplore and ACM), using the methodology for systematic review proposed by Kitchenham. Based on this methodology, we used tags and exclusion criteria to select papers about the topic under study. After the works selection, we applied a categorization process aiming to group similar object representation types, analyzing the steps applied for object recognition, the tests and evaluation performed and the databases used. Lastly, we compressed all the obtained information in a general overview and presented future prospects for the area.

关键词： 3d object recognition 3d object classification 3d object representation Systematic literature review

来源：评论

学校读者我要写书评

暂无评论

3d object recognition from range images using transform invariant object representation

引用

ELECTRONICS LETTERS 2010年第22期46卷 1499-U39页

作者： Akagunduz, E. Ulusoy, I. Middle E Tech Univ Elect & Elect Engn Dept TR-06531 Ankara Turkey

3d object recognition is performed using a scale and orientation invariant feature extraction method and a scale and orientation invariant topological representation. 3d surfaces are represented by sparse, repeatable, informative and semantically meaningful 3d surface structures, which are called multiscale features. These features are extracted with their scale (metric size and resolution) using the classified scale-space of 3d surface curvatures. Triplets of these features are used to represent the surface topologies invariant to transformation (scaling, rotation, translation) and geometric hashing is used for object recognition. Scaled and both scaled and occluded versions of range images from a 3d object database are tested and the experimental results are compared with other methods in the literature.

关键词： 3d object recognition 3d surface curvatures 3d surface representation feature extraction geometric hashing image classification image representation invariant topological representation multiscale features object recognition orientation invariant feature extraction method range images scale invariant feature extraction method scale-space classification transform invariant object representation transforms

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：