检索结果-内蒙古大学图书馆

Enhanced visual SLAM for construction robots by efficient integration of dynamic object segmentation and scene semantics

引用

ADVANCED ENGINEERING INFORMATICS 2024年 59卷

作者： Yang, Liu Cai, Hubo Purdue Univ Lyles Sch Civil Engn 550 Stadium Mall Dr W Lafayette IN 47907 USA

With the increasing adoption of autonomous mobile robots in the construction industry, accurate localization and mapping in dynamic construction environments have become paramount. This is typically tackled via Simultaneous Localization and Mapping (SLAM) techniques. Primarily designed for static environments, traditional SLAM systems struggle to maintain robustness and accuracy in dynamic settings. To address this challenge, this study presents an enhanced visual SLAM system specifically tailored for dynamic construction environments. The proposed system, named vSLAM-Con, introduces an adaptive dynamic object segmentation method, utilizing an innovative AD-keyframes selection mechanism grounded on optical flow magnitude to diminish computational overhead while preserving competitive tracking accuracy. Additionally, a semantic-based feature update process is developed, leveraging scene understanding and continuous observation to augment the reliability of tracking features. This system's performance, evaluated on both an established public benchmark and a custom construction dataset, shows substantial improvements over the baseline and competitive results with the state-of-the-art algorithms. More importantly, it largely reduces the processing time compared to state-of-the-arts, demonstrating robust tracking performance even under highly dynamic conditions. The findings highlight the system's potential to contribute significantly to autonomous robotics in construction, offering more accurate navigation and interaction capabilities in complex, ever-changing environments.

关键词： Visual simultaneous localization and mapping (SLAM) Construction environment dynamic object segmentation Scene semantics Structural elements

来源：评论

学校读者我要写书评

暂无评论

3D Lidar Point Cloud dynamic object segmentation Based on Multi-Hierarchical Feature Extraction

3D Lidar Point Cloud Dynamic Object Segmentation Based on Mu...

引用

第34届中国控制与决策会议

作者： Fang Li Fengshan Zou Zhenjun Du Faculty of Robots Science and Engineering Northeastern University SIASUN Robot & Automation CO. Ltd

Normally,SLAM algorithms are based on the assumption of static *** quality of lidar input directly affects the final performance of SLAM algorithm during this *** important factor that affects the input quality is the dynamic objects in the ***,how to segment dynamic objects from the 3 D point cloud to improve the quality of input is a question worth paying attention *** contrast to traditional methods,we use convolutional neural network to segment points on moving ***'s achallenge to improve the segmentation accuracy with a low amount of computation effort by carefully designing the network structure to control the number of parameters of the network.A method was proposed to convert the 3 D point cloud into a range image and combine the residual information as the input of CNN and the multi-hierarchical feature extraction module was used to improve the accuracy of the existing *** focus on segmenting moving objects rather than dividing the points into their semantic *** have shown that the proposed method can effectively improve the accuracy of dynamic object segmentation while keeping the neural network model lightweight.

关键词： 3D Lidar Point Cloud dynamic object segmentation Multi-Hierarchical Feature Extraction

来源：评论

学校读者我要写书评

暂无评论

DOS-SLAM: A Real-time dynamic object segmentation Visual SLAM System 19

DOS-SLAM: A Real-time Dynamic Object Segmentation Visual SLA...

引用

Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence

作者： He Xu Sheng Zhang Pengyu Liu Department of Electronic Engineering Tsinghua University Beijing China Advanced Sensor and Integrated System Lab Graduate School at Shenzhen Tsinghua University Shenzhen China Institute of Microelectronics Tsinghua University Beijing China

ISBN: (纸本)9781450372619

SLAM technology has developed very rapidly in recent years, and systems for general environment has achieved excellent results. However, there are still some practical problems affecting the robustness and accuracy of SLAM, including the impact of highly dynamic environments. In this paper, we propose a real-time SLAM system, named DOS-SLAM, which is specifically designed for dynamic environments. The system, based on ORB-SLAM2, adds two modules to achieve accurate in highly dynamic environments, dynamic object segmentation with instance segmentation and geometric constraint and static semantic octo-tree map to realize a static map without moving objects. TUM RGB-D Dataset is used to present the performance of DOS-SLAM, and it has been proved that DOS-SLAM has outstanding and superior performance in dynamic environment. Meanwhile, it can realize real-time computing. In addition, static semantics octo-tree map built by DOS-SLAM system occupies less memory and only contains static objects, which is more conducive to robot navigation and large-scale scene map generation.

关键词： Visual SLAM Static semantic map Real-time dynamic object segmentation

来源：评论

学校读者我要写书评

暂无评论

DOS-SLAM: A Real-time dynamic object segmentation Visual SLAM System

DOS-SLAM: A Real-time Dynamic Object Segmentation Visual SLA...

引用

2019第二届算法、计算和人工智能国际会议

作者： He Xu Sheng Zhang Pengyu Liu Department of Electronic Engineering Tsinghua University Advanced Sensor and Integrated System Lab Graduate School at Shenzhen Tsinghua University Institute of Microelectronics Tsinghua University

SLAM technology has developed very rapidly in recent years, and systems for general environment has achieved excellent results. However, there are still some practical problems affecting the robustness and accuracy of SLAM, including the impact of highly dynamic environments. In this paper, we propose a real-time SLAM system, named DOS-SLAM, which is specifically designed for dynamic environments. The system, based on ORBSLAM2, adds two modules to achieve accurate in highly dynamic environments, dynamic object segmentation with instance segmentation and geometric constraint and static semantic octotree map to realize a static map without moving objects. TUM RGB-D Dataset is used to present the performance of DOSSLAM, and it has been proved that DOS-SLAM has outstanding and superior performance in dynamic environment. Meanwhile, it can realize real-time computing. In addition, static semantics octotree map built by DOS-SLAM system occupies less memory and only contains static objects, which is more conducive to robot navigation and large-scale scene map generation.

关键词： dynamic object segmentation Static semantic map Real-time Visual SLAM

来源：评论

学校读者我要写书评

暂无评论

A novel vSLAM framework with unsupervised semantic segmentation based on adversarial transfer learning

引用

APPLIED SOFT COMPUTING 2020年 90卷

作者： Jin, Sheng Chen, Liang Sun, Rongchuan McLoone, Sean Soochow Univ Sch Mech & Elect Engn Suzhou 215021 Peoples R China Queens Univ Belfast Sch Elect Elect Engn & Comp Sci Belfast Antrim North Ireland

Significant progress has been made in the field of visual Simultaneous Localization and Mapping (vSLAM) systems. However, the localization accuracy of vSLAM can be significantly reduced in dynamic applications with mobile robots or passengers. In this paper, a novel semantic SLAM framework in dynamic environments is proposed to improve the localization accuracy. We incorporate a semantic segmentation model into the Oriented FAST and Rotated BRIEF-SLAM2 (ORB-SLAM2) system to filter out dynamic feature points, but we encounter one main challenge, i.e. the performance of a segmentation network well-trained with labeled datasets may decrease seriously in a real application without any labeled data due to the inconsistency between the source domain and the target domain. Therefore, we proposed an unsupervised semantic segmentation model with a Residual Neural Network (ResNet) structure, which is trained by the adversarial transfer learning method in the multi-level feature spaces. This work may be the first to perform multi-level feature space adversarial transfer learning for the semantic SLAM task in dynamic environments. In order to evaluate our method, images of indoor scenes from three datasets are used as the source domain, and the dynamic sequences of the TUM dataset are used as the target domain. The extensive experimental results show favorable performance against the state-of-the-art methods in terms of the absolute trajectory accuracy and image semantic segmentation quality. (C) 2020 Elsevier B.V. All rights reserved.

关键词： Semantic SLAM dynamic environments dynamic object segmentation

来源：评论

学校读者我要写书评

暂无评论

Visual SLAM and Structure from Motion in dynamic Environments: A Survey

引用

ACM COMPUTING SURVEYS 2018年第2期51卷 1–36页

作者： Saputra, Muhamad Risqi U. Markham, Andrew Trigoni, Niki Univ Oxford Dept Comp Sci Wolfson BldgParks Rd Oxford OX1 3QD England

In the last few decades, Structure from Motion (SfM) and visual Simultaneous Localization and Mapping (visual SLAM) techniques have gained significant interest from both the computer vision and robotic communities. Many variants of these techniques have started to make an impact in a wide range of applications, including robot navigation and augmented reality. However, despite some remarkable results in these areas, most SfM and visual SLAM techniques operate based on the assumption that the observed environment is static. However, when faced with moving objects, overall system accuracy can be jeopardized. In this article, we present for the first time a survey of visual SLAM and SfM techniques that are targeted toward operation in dynamic environments. We identify three main problems: how to perform reconstruction (robust visual SLAM), how to segment and track dynamic objects, and how to achieve joint motion segmentation and reconstruction. Based on this categorization, we provide a comprehensive taxonomy of existing approaches. Finally, the advantages and disadvantages of each solution class are critically discussed from the perspective of practicality and robustness.

关键词： Structure from motion visual SLAM deep learning 3D reconstruction visual odometry motion segmentation dynamic object segmentation 3D tracking dynamic environments

来源：评论

学校读者我要写书评

暂无评论

A model for dynamic object segmentation with kernel density estimation based on gradient features

引用

IMAGE AND VISION COMPUTING 2009年第6期27卷 817-823页

作者： Li, Zheng Jiang, Pohuang Ma, Hong Yang, Jian Tang, DongMing Sichuan Univ Coll Comp Sci Chengdu 610064 Sichuan Peoples R China Sichuan Univ Coll Math Chengdu 610064 Sichuan Peoples R China Univ Elect Sci & Technol China Chengdu 610054 Sichuan Peoples R China

The dynamic object segmentation in videos taken from a static camera is a basic technique in many vision surveillance applications. In order to suppress fake objects caused by dynamic cast shadows and reflection images, this paper presents a novel segmentation model with the function of cast shadow and reflection image suppression. This model is a kernel density estimation model based on dynamic gradient features. Unlike the conventional kernel density estimation model which can only suppress cast shadows in color videos, this model Can also Suppress them in intensity videos, and under the circumstance of diffusion it can suppress reflection images effectively. Although this model may cause the increase of the false negative rate, its function of fake object Suppression is remarkable. Furthermore, the false negative rate can be reduced with other convenient methods. Some experimental results by real videos are also presented in this paper to demonstrate the effectiveness of this model. (C) 2008 Elsevier B.V. All rights reserved.

关键词： Kernel density estimation Cast shadow Reflection image Gradient feature dynamic object segmentation

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：