检索结果-内蒙古大学图书馆

2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC)

作者： Liu, Yaohua Ma, Yue Xu, Min Chinese Acad Sci Shenyang Inst Automat Shenyang 110016 Peoples R China Chinese Acad Sci Inst Robot Shenyang 110169 Peoples R China Chinese Acad Sci Inst Intelligent Mfg Shenyang 110169 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Key Lab Intelligent Detect & Equipment Technol Li Shenyang 100179 Peoples R China

ISBN: (纸本)9781510655201;9781510655195

Point cloud can represent 3d geometry conveniently, but its challenging for computers to process it. In this work, we design a transformer enhanced hierarchical neural network for accurate large scale point cloud semantic segmentation. We use semantic space's transformer block to learn global feature correlation. In this way, we can expand the receptive field of network to the whole input point cloud. Experimental results on S3dIS 3d semantic segmentation dataset show that, compared with the traditional hierarchical 3d semantic segmentation model, our transformer-enhanced hierarchical model achieved higher performance on overall accuracy and mIoU.

关键词： Transformer hierarchical 3d semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

semantic segmentation of 3d point cloud data acquired from robot dog for scaffold monitoring 38

Semantic segmentation of 3D point cloud data acquired from r...

引用

38th International Symposium on Automation and Robotics in Construction, ISARC 2021

作者： Kim, Juhyeon Chung, duho Kim, Yohan Kim, Hyoungkwan Department of Civil and Environmental Engineering Yonsei University Korea Republic of

ISBN: (纸本)9789526952413

Many of the fatalities and injuries in the construction industry occur in scaffolding accidents, and monitoring the scaffolding process and checking compliance are critical. However, monitoring scaffolds is labor-intensive and inefficient because it is done manually. To address this issue, we propose an advanced 3d reconstruction method for detecting and monitoring scaffolds. deep learning-based RandLA-Net architecture is used to perform scene segmentation. RandLA-Net is trained based on transfer learning, using the knowledge of the model learned with the semantic3d dataset. RandLA-Net uses 3d point cloud data that are matched and registered by LIO-SAM, a laser slam algorithm. By attaching a LidAR to a quadruped robot, it is possible to obtain data frequently in a manner suitable for construction sites. The proposed methodology has demonstrated good performance in monitoring scaffolds. © 2021 Proceedings of the International Symposium on Automation and Robotics in Construction. All rights reserved.

关键词： 3d semantic segmentation Mobile Laser Scanning (MLS) Robot dog Scaffold Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

Weakly supervised 3d point cloud semantic segmentation for architectural heritage using teacher-guided consistency and contrast learning

引用

AUTOMATION IN CONSTRUCTION 2024年 168卷

作者： Huang, Shuowen Hu, Qingwu Ai, Mingyao Zhao, Pengcheng Li, Jian Cui, Hao Wang, Shaohua Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430079 Peoples R China Zhengzhou Univ Sch Geosci & Technol Zhengzhou 450001 Peoples R China

Point cloud semantic segmentation is significant for managing and protecting architectural heritage. Currently, fully supervised methods require a large amount of annotated data, while weakly supervised methods are difficult to transfer directly to architectural heritage. This paper proposes an end-to-end teacher-guided consistency and contrastive learning weakly supervised (TCCWS) framework for architectural heritage point cloud semantic segmentation, which can fully utilize limited labeled points to train network. Specifically, a teacherstudent framework is designed to generate pseudo labels and a pseudo label dividing module is proposed to distinguish reliable and ambiguous point sets. Based on it, a consistency and contrastive learning strategy is designed to fully utilize supervision signals to learn the features of point clouds. The framework is tested on the ArCH dataset and self-collected point cloud, which demonstrates that the proposed method can achieve effective semantic segmentation of architectural heritage using only 0.1 % of annotated points.

关键词： Point cloud Architectural heritage 3d semantic segmentation Weakly supervised

来源：评论

学校读者我要写书评

暂无评论

3d-based precise evaluation pipeline for maize ear rot using multi-view stereo reconstruction and point cloud semantic segmentation

引用

COMPUTERS ANd ELECTRONICS IN AGRICULTURE 2024年 216卷

作者： Yang, Rui He, Yong Lu, Xiangyu Zhao, Yiying Li, Yanmei Yang, Yinhui Kong, Wenwen Liu, Fei Zhejiang Univ Coll Biosyst Engn & Food Sci Hangzhou 310058 Peoples R China Zhejiang Acad Agr Sci Inst Digital Agr Hangzhou 310021 Peoples R China China Agr Univ Natl Maize Improvement Ctr China Beijing 100193 Peoples R China Zhejiang A&F Univ Coll Math & Comp Sci Hangzhou 311300 Peoples R China

Maize ear rot poses a severe threat to maize yield and quality. Breeding and cultivating highly resistant maize varieties is a crucial approach for preventing and controlling maize ear rot. However, traditional methods of visually grading the severity of maize ear infection and resistance lack objectivity and repeatability. To meet the requirement of precise breeding and resistance assessment scenarios, a novel pipeline based on threedimensional (3d) point clouds of maize ear was developed for ear rot precise evaluation. First, multi-view stereo (MVS) reconstruction was employed to obtain high-precision dense point clouds of maize ears. And the coordinate correction and circular sampling approaches were proposed to optimize the data structure of the input maize ear samples. Next, a specialized network called the ear rot segmentation network (ERSegNet) was proposed to detect the infected area of maize ears. This network incorporated an orientation-encoding (OE) module and point transformer (PT) attention, which effectively boosted the performance of PointNet++. The proposed ERSegNet achieved impressive results, including a mean intersection over union (mIoU) of 85.83%, a mean precision (mPrec) of 92.34%, a mean recall (mRec) of 92.23%, a mean F1-score of 92.28%, and an overall accuracy (OA) of 93.76%. This demonstrated the feasibility of using semantic segmentation algorithms to predict 3d point clouds of maize ears. Furthermore, a point cloud resampling method was suggested to enhance the spatial uniformity of maize ear point clouds and a point-level quantitative assessment approach based on the 3d point cloud data was provided for evaluating the severity of ear rot. The results showed an average evaluation error of 1.55% in the testing set, indicating the accuracy of the proposed method. This study provides a reliable and objective method for maize ear rot precise assessment, offering potential and valuable support for the identification of resistant varieties in b

关键词： Maize ear rot Multi-view stereo reconstruction Point cloud 3d semantic segmentation deep learning

来源：评论

学校读者我要写书评

暂无评论

Swin3d: A pretrained transformer backbone for 3d indoor scene understanding

引用

Computational Visual Media 2025年第1期11卷 83-101页

作者： Yu-Qi Yang Yu-Xiao Guo Jian-Yu Xiong Yang Liu Hao Pan Peng-Shuai Wang Xin Tong Baining Guo Institute for Advanced Study Tsinghua UniversityBeijing 100084China Tsinghua Shenzhen International Graduate School Tsinghua UniversityShenzhen 518055China Internet Graphics Group Microsoft Research AsiaBeijing 100080China Wangxuan Institute of Computer Technology Peking UniversityBeijing 100080China

The use of pretrained backbones with finetuning has shown success for 2d vision and natural language processing tasks,with advantages over taskspecific *** this paper,we introduce a pretrained 3d backbone,called Swin3d,for 3d indoor scene *** designed a 3d Swin Transformer as our backbone network,which enables efficient selfattention on sparse voxels with linear memory complexity,making the backbone scalable to large models and *** also introduce a generalized contextual relative positional embedding scheme to capture various irregularities of point signals for improved network *** pretrained a large Swin3d model on a synthetic Structured3d dataset,which is an order of magnitude larger than the ScanNet *** model pretrained on the synthetic dataset not only generalizes well to downstream segmentation and detection on real 3d point datasets but also outperforms state-of-the-art methods on downstream tasks with+2.3 mIoU and+2.2 mIoU on S3dIS Area5 and 6-fold semantic segmentation,respectively,+1.8 mIoU on ScanNet segmentation(val),+1.9 mAP@0.5 on ScanNet detection,and+8.1 mAP@0.5 on S3dIS detection.A series of extensive ablation studies further validated the scalability,generality,and superior performance enabled by our approach.

关键词： 3d pretraining ponitcloud analysis trans-former backbone Swin Transformer 3d semantic segmentation 3d object detection

来源：评论

学校读者我要写书评

暂无评论

A point cloud segmentation network with hybrid convolution and differential channels

引用

SCIENTIFIC REPORTS 2025年第1期15卷 1-13页

作者： Zhang, Xiaoyan Bu, Yantao Xian Univ Sci & Technol Comp Sci & Technol Coll Xian 710600 Peoples R China

In recent years, point-based segmentation methods have made significant progress in improving segmentation accuracy. However, existing approaches still suffer from several key limitations. Traditional convolution operations make it difficult to effectively model the irregular geometry in point cloud data, resulting in insufficient sensitivity to spatial details. Additionally, current methods have limitations in collaboratively modeling and integrating global and local information. For this reason, we propose a 3d segmentation network based on hybrid convolution and differential channels. Specifically, we design a hybrid convolutional feature extraction (HCFE) module for processing 3d semantic information and spatial information independently, using different convolution kernels to obtain the subtle geometric structure differences between points. Then, we propose a differential Channel Feature Interaction (dCFI) Module to enhance the local details and global channel information through differential Convolution (dCU) and a Simplified Channel Attention Mechanism (S_ECA), respectively, and adaptively fuse the two types of information by dynamic Interaction Mechanism (dIM), achieving their cooperative optimization. Compared to existing methods, HdC_Net has clear advantages in detail capturing and the integration of local and global information. A large number of experiments demonstrate the effectiveness and superiority of the model proposed in this study.

关键词： differential convolution 3d semantic segmentation 3d part segmentation deep learning

来源：评论

学校读者我要写书评

暂无评论

Uni-to-Multi Modal Knowledge distillation for Bidirectional LidAR-Camera semantic segmentation

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS ANd MACHINE INTELLIGENCE 2024年第12期46卷 11059-11072页

作者： Sun, Tianfang Zhang, Zhizhong Tan, Xin Peng, Yong Qu, Yanyun Xie, Yuan East China Normal Univ Sch Comp Sci & Technol Shanghai 200060 Peoples R China Shanghai Key Lab Comp Software Evaluating & Testi Shanghai 200240 Peoples R China Cent South Univ Sch Traff & Transportat Engn Changsha 410017 Peoples R China Xiamen Univ Dept Comp Sci & Technol Xiamen 361005 Peoples R China

Combining LidAR points and images for robust semantic segmentation has shown great potential. However, the heterogeneity between the two modalities (e.g. the density, the field of view) poses challenges in establishing a bijective mapping between each point and pixel. This modality alignment problem introduces new challenges in network design and data processing for cross-modal methods. Specifically, 1) points that are projected outside the image planes;2) the complexity of maintaining geometric consistency limits the deployment of many data augmentation techniques. To address these challenges, we propose a cross-modal knowledge imputation and transition approach. First, we introduce a bidirectional feature fusion strategy that imputes missing image features and performs cross-modal fusion simultaneously. This allows us to generate reliable predictions even when images are missing. Second, we propose a Uni-to-Multi modal Knowledge distillation (U2MKd) framework, leveraging the transfer of informative features from a single-modality teacher to a cross-modality student. This overcomes the issues of augmentation misalignment and enables us to train the student effectively. Extensive experiments on the nuScenes, Waymo, and semanticKITTI datasets demonstrate the effectiveness of our approach. Notably, our method achieves an 8.3 mIoU gain over the LidAR-only baseline on the nuScenes validation set and achieves state-of-the-art performance on the three datasets.

关键词： 3d semantic segmentation cross-modal feature imputing cross-modal fusion cross-modal knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

Meta-RangeSeg: LidAR Sequence semantic segmentation Using Multiple Feature Aggregation

引用

IEEE ROBOTICS ANd AUTOMATION LETTERS 2022年第4期7卷 9739-9746页

作者： Wang, Song Zhu, Jianke Zhang, Ruixiang Zhejiang Univ Coll Comp Sci Hangzhou 310027 Peoples R China Alibaba Zhejiang Univ Joint Res Inst Frontier Tec Hangzhou 310027 Peoples R China Hikvis Res Inst Hangzhou 310051 Peoples R China

LidAR sensor is essential to the perception system in autonomous vehicles and intelligent robots. To fulfill the real-time requirements in real-world applications, it is necessary to efficiently segment the LidAR scans. Most of previous approaches directly project 3d point cloud onto the 2d spherical range image so that they can make use of the efficient 2d convolutional operations for image segmentation. Although having achieved the encouraging results, the neighborhood information is not well-preserved in the spherical projection. Moreover, the temporal information is not taken into consideration in the single scan segmentation task. To tackle these problems, we propose a novel approach to semantic segmentation for LidAR sequences named Meta-RangeSeg, where a new range residual image representation is introduced to capture the spatial-temporal information. Specifically, Meta-Kernel is employed to extract the meta features, which reduces the inconsistency between the 2d range image coordinates input and 3d Cartesian coordinates output. An efficient U-Net backbone is used to obtain the multi-scale features. Furthermore, Feature Aggregation Module (FAM) strengthens the role of range channel and aggregates features at different levels. We have conducted extensive experiments for performance evaluation on semanticKITTI and semanticPOSS. The promising results show that our proposed Meta-RangeSeg method is more efficient and effective than the existing approaches.

关键词： 3d semantic segmentation LidAR perception autonomous vehicle

来源：评论

学校读者我要写书评

暂无评论

Live semantic 3d Perception for Immersive Augmented Reality

引用

IEEE TRANSACTIONS ON VISUALIZATION ANd COMPUTER GRAPHICS 2020年第5期26卷 2012-2022页

作者： Han, Lei Zheng, Tian Zhu, Yinheng Xu, Lan Fang, Lu Tsinghua Univ Beijing Peoples R China Hong Kong Univ Sci & Technol Hong Kong Peoples R China

semantic understanding of 3d environments is critical for both the unmanned system and the human involved virtual/augmented reality (VR/AR) immersive experience. Spatially-sparse convolution, taking advantage of the intrinsic sparsity of 3d point cloud data, makes high resolution 3d convolutional neural networks tractable with state-of-the-art results on 3d semantic segmentation problems. However, the exhaustive computations limits the practical usage of semantic 3d perception for VR/AR applications in portable devices. In this paper, we identify that the efficiency bottleneck lies in the unorganized memory access of the sparse convolution steps, i.e., the points are stored independently based on a predefined dictionary, which is inefficient due to the limited memory bandwidth of parallel computing devices (GPU). With the insight that points are continuous as 2d surfaces in 3d space, a chunk-based sparse convolution scheme is proposed to reuse the neighboring points within each spatially organized chunk. An efficient multi-layer adaptive fusion module is further proposed for employing the spatial consistency cue of 3d data to further reduce the computational burden. Quantitative experiments on public datasets demonstrate that our approach works 11x faster than previous approaches with competitive accuracy. By implementing both semantic and geometric 3d reconstruction simultaneously on a portable tablet device, we demo a foundation platform for immersive AR applications.

关键词： Three-dimensional displays Convolution semantics Two dimensional displays Image segmentation Graphics processing units Solid modeling dense 3d Reconstruction 3d semantic segmentation 3d Convolutional Network Virtual Reality Augmented Reality

来源：评论

学校读者我要写书评

暂无评论

3d multi-scale feature extraction and recalibration network for spinal structure and lesion segmentation

引用

ACTA RAdIOLOGICA 2023年第12期64卷 3015-3023页

作者： Wang, Hongjie Chen, Yingjin Jiang, Tao Bian, Huwei Shen, Xing Nanjing Univ Aeronaut & Astronaut State Key Lab Mech & Control Mech Struct Nanjing Peoples R China Changzhou Tradit Chinese Med Hosp Dept Orthopaed Changzhou Jiangsu Peoples R China

Background Automatic segmentation has emerged as a promising technique for the diagnosis of spinal conditions. Purpose To design and evaluate a deep convolution network for segmenting the intervertebral disc, spinal canal, facet joint, and herniated disk on magnetic resonance imaging (MRI) scans. Material and Methods MRI scans of 70 patients with disc herniation were gathered and manually annotated by radiologists. A novel deep neural network was developed, comprising 3d squeeze-and-excitation blocks and multi-scale feature extraction blocks for automated segmentation of spinal structure and lesion. To address the issue of class imbalance, a weighted cross-entropy loss was introduced for training. In addition, semi-supervision segmentation was accomplished to reduce annotation labor cost. Results The proposed model achieved 77.67% mean intersection over union, with 9.56% and 11.11% gains over typical V-Net and U-Net respectively, outperforming the other models in ablation experiments. In addition, the semi-supervision segmentation method was proven to work. Conclusion The 3d multi-scale feature extraction and recalibration network achieved an excellent segmentation performance of intervertebral disc, spinal canal, facet joint, and herniated disk, outperforming typical encoder-decoder networks.

关键词： deep learning 3d semantic segmentation lumbar spine magnetic resonance imaging 3d V-Net SE block atrous convolution

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：