检索结果-内蒙古大学图书馆

Attention-guided LidAR segmentation and odometry using image-to-point cloud saliency transfer

MULTIMEdIA SYSTEMS 2024年第4期30卷 188-188页

作者： ding, Guanqun Imamoglu, Nevrez Caglayan, Ali Murakawa, Masahiro Nakamura, Ryosuke Univ Tsukuba Grad Sch Sci & Technol Tsukuba Ibaraki 3058577 Japan Natl Inst Adv Ind Sci & Technol Digital Architecture Res Ctr DigiARC Tokyo 1350064 Japan Natl Inst Adv Ind Sci & Technol CNRS AIST Joint Robot Lab JRL Tsukuba Ibaraki 3058560 Japan Natl Inst Adv Ind Sci & Technol Artificial Intelligence Res Ctr AIRC Tokyo 1350064 Japan

LidAR odometry estimation and 3d semantic segmentation are crucial for autonomous driving, which has achieved remarkable advances recently. However, these tasks are challenging due to the imbalance of points in different semantic categories for 3d semantic segmentation and the influence of dynamic objects for LidAR odometry estimation, which increases the importance of using representative/salient landmarks as reference points for robust feature learning. To address these challenges, we propose a saliency-guided approach that leverages attention information to improve the performance of LidAR odometry estimation and semantic segmentation models. Unlike in the image domain, only a few studies have addressed point cloud saliency information due to the lack of annotated training data. To alleviate this, we first present a universal framework to transfer saliency distribution knowledge from color images to point clouds, and use this to construct a pseudo-saliency dataset (i.e. FordSaliency) for point clouds. Then, we adopt point cloud based backbones to learn saliency distribution from pseudo-saliency labels, which is followed by our proposed SalLidAR module. SalLidAR is a saliency-guided 3d semantic segmentation model that integrates saliency information to improve segmentation performance. Finally, we introduce SalLONet, a self-supervised saliency-guided LidAR odometry network that uses the semantic and saliency predictions of SalLidAR to achieve better odometry estimation. Our extensive experiments on benchmark datasets demonstrate that the proposed SalLidAR and SalLONet models achieve state-of-the-art performance against existing methods, highlighting the effectiveness of image-to-LidAR saliency knowledge transfer. Source code will be available at https://***/nevrez/SalLONet

关键词： LidAR odometry estimation Saliency detection 3d semantic segmentation Point cloud

来源：评论

学校读者我要写书评

暂无评论

Enhancing Autonomous Train Safety Through A Priori-Map Based Perception 4th

Enhancing Autonomous Train Safety Through A Priori-Map Based...

引用

4th International Conference on Reliability, Safety, and Security of Railway Systems (RSSRail)

作者： Mahtani, Ankur Chouchani, Nadia Herbreteau, Maxime Rafin, denis FCS Railenium 180 Rue Joseph Louis Lagrange F-59300 Famars France Thales Serv Numer SAS 290 Allee Lac F-31670 Labege France SpirOps 8 Passage Bonne Graine F-75011 Paris France

ISBN: (纸本)9783031058134;9783031058141

Autonomous driving tends to increase use of perception as a tool for analyzing the environment before making a decision that could impact driving. However, recent techniques based on machine learning do not provide the necessary interpretability to ensure sufficient driving safety. Combining multiple sources, deterministic or not, allows results to be cross-referenced and therefore more reliable. In this paper, we propose a novel methodology that aligns an infrastructure mapping system and point cloud analysis for railway tracks and catenaries perception to ensure autonomous train's safety. By using a deep learning model to recognize and classify rails with the implicit knowledge of the railway infrastructure, we exceed in performance all previous systems of infrastructure: 60.9% in mIoU for tracks segmentation and 9.27 points mMink for points alignment with ground-truth, at an interesting runtime of 20 Hz. Moreover, we propose an embedded solution for automatic monitoring which avoids hours of maintenance traffic on the railway tracks. This solution is used as acquisition system feeding map and perception in real-world data for autonomous trains.

关键词： Autonomous train Railway map Point cloud 3d semantic segmentation deep learning

来源：评论

学校读者我要写书评

暂无评论

data Efficient 3d Learner via Knowledge Transferred from 2d Model 1

引用

17th European Conference on Computer Vision (ECCV)

作者： Yu, Ping-Chung Sun, Cheng Sun, Min Natl Tsing Hua Univ Hsinchu Taiwan

ISBN: (数字)9783031198182

ISBN: (纸本)9783031198175;9783031198182

Collecting and labeling the registered 3d point cloud is costly. As a result, 3d resources for training are typically limited in quantity compared to the 2d images counterpart. In this work, we deal with the data scarcity challenge of 3d tasks by transferring knowledge from strong 2d models via RGB-d images. Specifically, we utilize a strong and well-trained semantic segmentation model for 2d images to augment RGB-d images with pseudo-label. The augmented dataset can then be used to pre-train 3d models. Finally, by simply fine-tuning on a few labeled 3d instances, our method already outperforms existing state-of-the-art that is tailored for 3d label efficiency. We also show that the results of mean-teacher and entropy minimization can be improved by our pre-training, suggesting that the transferred knowledge is helpful in semi-supervised setting. We verify the effectiveness of our approach on two popular 3d models and three different tasks. On ScanNet official evaluation, we establish new state-of-the-art semantic segmentation results on the data-efficient track.

关键词： Knowledge transfer 3d semantic segmentation Point cloud recognition 3d pre-training Label efficiency

来源：评论

学校读者我要写书评

暂无评论

AMLP-Conv, a 3d Axial Long-range Interaction Multilayer Perceptron for CNNs 13th

AMLP-Conv, a 3D Axial Long-range Interaction Multilayer Perc...

引用

13th International Workshop on Machine Learning in Medical Imaging (MLMI)

作者： Bonheur, Savinien Pienn, Michael Olschewski, Horst Bischof, Horst Urschler, Martin Ludwig Boltzmann Inst Lung Vasc Res Graz Austria Graz Univ Technol Inst Comp Graph & Vis Graz Austria Med Univ Graz Dept Internal Med Graz Austria Univ Auckland Sch Comp Sci Auckland New Zealand

ISBN: (纸本)9783031210136;9783031210143

While Convolutional neural networks (CNN) have been the backbone of medical image analysis for years, their limited long-range interaction restrains their ability to encode long distance anatomical relationships. On the other hand, the current approach to capture long distance relationships, Transformers, is constrained by their quadratic scaling and their data inefficiency (arising from their lack of inductive biases). In this paper, we introduce the 3d Axial Multilayer Perceptron (AMLP), a long-range interaction module whose complexity scales linearly with spatial dimensions. This module is merged with CNNs to form the AMLP-Conv module, a long-range augmented convolution with strong inductive biases. Once combined with U-Net, our AMLP-Conv module leads to significant improvement, outperforming most transformer based U-Nets on the ACdC dataset, and reaching a new state-of-the-art result on the Multi-Modal Whole Heart segmentation (MM-WHS) dataset with an almost 1.1% dice score improvement over the previous scores on the Computed Tomography (CT) modality.

关键词： MLP Axial attention Convolutional neural network Multi-label 3d semantic segmentation Heart segmentation

来源：评论

学校读者我要写书评

暂无评论

4dContrast: Contrastive Learning with dynamic Correspondences for 3d Scene Understanding 17th

4DContrast: Contrastive Learning with Dynamic Correspondence...

引用

17th European Conference on Computer Vision (ECCV)

作者： Chen, Yujin Niessner, Matthias dai, Angela Tech Univ Munich Munich Germany

ISBN: (纸本)9783031198236;9783031198243

We present a new approach to instill 4d dynamic object priors into learned 3d representations by unsupervised pre-training. We observe that dynamic movement of an object through an environment provides important cues about its objectness, and thus propose to imbue learned 3d representations with such dynamic understanding, that can then be effectively transferred to improved performance in downstream 3d semantic scene understanding tasks. We propose a new data augmentation scheme leveraging synthetic 3d shapes moving in static 3d environments, and employ contrastive learning under 3d-4d constraints that encode 4d invariances into the learned 3d representations. Experiments demonstrate that our unsupervised representation learning results in improvement in downstream 3d semantic segmentation, object detection, and instance segmentation tasks, and moreover, notably improves performance in data-scarce scenarios. Our results show that our 4d pretraining method improves downstream tasks such as object detection mAP@0.5 by 5.5%/6.5% over training from scratch on ScanNet/SUN RGB-d while involving no additional run-time overhead at test time.

关键词： 3d scene understanding Point cloud recognition 3d semantic segmentation 3d instance segmentation 3d object detection

来源：评论

学校读者我要写书评

暂无评论

Automated bridge component recognition using close-range images from unmanned aerial vehicles

引用

ENGINEERING STRUCTURES 2023年 274卷 1页

作者： Kim, Hyunjun Narazaki, Yasutaka Spencer Jr, Billie F. Seoul Natl Univ Sci & Technol Dept Civil Engn Seoul 01811 South Korea Zhejiang Univ Univ Illinois Urbana Champaign Inst Haining 314400 Peoples R China Univ Illinois Dept Civil & Environm Engn Urbana IL 61801 USA

Unmanned aerial vehicles (UAVs), in conjunction with computer vision techniques, have shown great potential for bridge inspections. Close-range images captured in proximity to the structural surface are generally required to detect damage and also need to be linked to the corresponding structural component to enable assessment of the health of the global structure. However, the lack of contextual information makes automated identification of bridge components in close-range images challenging. This study proposes a framework for automated bridge component recognition based on close-range images collected by UAVs. First, a 3d point cloud is generated from the UAV survey of the bridge and segmented into bridge components. The segmented point cloud is subsequently projected onto the camera coordinates to categorize each of the images into the bridge component. The proposed approach is successfully validated using a local highway bridge, pointing the way for improved inspection of full-scale bridges.

关键词： 3d semantic segmentation Automated structural inspection Bridge components Close-range images Computer vision Point cloud Unmanned aerial vehicle (UAV)

来源：评论

学校读者我要写书评

暂无评论

3d reconstruction of large-scale scaffolds with synthetic data generation and an upsampling adversarial network

引用

AUTOMATION IN CONSTRUCTION 2023年 156卷

作者： Kim, Juhyeon Kim, Jeehoon Kim, Yohan Kim, Hyoungkwan Yonsei Univ Sch Civil & Environm Engn Seoul South Korea

Falls from scaffolds cause the majority of accidents and fatalities at construction sites. A deep learning-based 3d reconstruction technology could provide a solution to prevent such fatalities through automated scaffold monitoring. However, when the technology was used at a large-scale construction site, there were limitations, such as the scarcity of point cloud data and the non-uniformity of points. To address this issue, this paper presents a large-scale scaffold reconstruction method using synthetic scaffold datasets and an upsampling adversarial network. The method consists of four steps: 1) data acquisition of scaffold point cloud through a mobile laser scanning (MLS) system, 2) 3d semantic segmentation using synthetic datasets, 3) upsampling of the segmented scaffold points, and 4) automatic generation of a 3d CAd model. The performance of the segmentation model trained with synthetic datasets achieved an 80.83% F1 score, which improved to 94.93% after upsampling.

关键词： Mobile laser scanning Scaffold monitoring Synthetic dataset Upsampling adversarial network 3d point cloud data 3d semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

A fast end-to-end method for automatic interior progress evaluation using panoramic images

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年第PartA期126卷

作者： Fang, Xin Li, Heng Wu, Haitao Fan, Lang Kong, Ting Wu, Yue Hong Kong Polytech Univ Dept Bldg & Real Estate Hung Hom Kowloon Hong Kong 999077 Peoples R China

Interior construction makes up a large portion of project budget and time and is more prone to schedule delays. Most research efforts on progress management focus on exterior environment, while few on interior construction. Although progress monitoring methods based on laser point clouds and computer vision are investigated before, the problems of costly acquisition and creation of point clouds and images are still open, which impede the study of progress evaluation, particularly in interior construction environments where clutters and occlusions are universal. This paper introduces a method based on 360 degrees panoramic images and deep learning for fast end-to-end interior progress evaluation in room units. The method takes only one or two 360 degrees panoramic images as input, estimates key corners, generates and registers room layouts, and semantically segments sparse point cloud. With the extracted corners and segmentation results, the progress states of interior trades can be evaluated. The experimental results show that the proposed method based on deep learning techniques achieves comparable performance against those on public data sets with 3d Intersection over Union (3d IoU) of 83.69% vs 84.23%, Corner Error (CE) of 0.4% vs 0.69%, and mean class Interaction over Union (mIoU) of 70.28% vs 53.5%. A case study of an interior decoration project of a hotel is adopted to demonstrate the feasibility and practical capabilities of the proposed method.

关键词： Interior progress evaluation Room layout reconstruction 3d registration 3d semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Mix3d: Out-of-Context data Augmentation for 3d Scenes 9

Mix3D: Out-of-Context Data Augmentation for 3D Scenes

引用

9th International Conference on 3d Vision (3dV)

作者： Nekrasov, Alexey Schult, Jonas Litany, Or Leibe, Bastian Engelmann, Francis Rhein Westfal TH Aachen Aachen Germany NVIDIA Sunnyvale CA USA ETH AI Ctr Zurich Switzerland

ISBN: (纸本)9781665426886

We present Mix3d, a data augmentation technique for segmenting large-scale 3d scenes. Since scene context helps reasoning about object semantics, current works focus on models with large capacity and receptive fields that can fully capture the global context of an input 3d scene. However, strong contextual priors can have detrimental implications like mistaking a pedestrian crossing the street for a car. In this work, we focus on the importance of balancing global scene context and local geometry, with the goal of generalizing beyond the contextual priors in the training set. In particular, we propose a "mixing" technique which creates new training samples by combining two augmented scenes. By doing so, object instances are implicitly placed into novel out-of-context environments, therefore making it harder for models to rely on scene context alone, and instead infer semantics from local structure as well. We perform detailed analysis to understand the importance of global context, local structures and the effect of mixing scenes. In experiments, we show that models trained with Mix3d profit from a significant performance boost on indoor (ScanNet, S3dIS) and outdoor datasets (semanticKITTI). Mix3d can be trivially used with any existing method, e.g., trained with Mix3d, MinkowskiNet outperforms all prior state-of-the-art methods by a significant margin on the ScanNet test benchmark (78.1% mIoU). Code is available at: https://***/mix3d/

关键词： 3d data augmentation 3d scene understanding 3d semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

deep learning-based 3d reconstruction of scaffolds using a robot dog

引用

AUTOMATION IN CONSTRUCTION 2022年第0期134卷 104092-104092页

作者： Kim, Juhyeon Chung, duho Kim, Yohan Kim, Hyoungkwan Yonsei Univ Sch Civil & Environm Engn Seoul South Korea

Although a scaffold is an essential structure in the construction industry, it may also be a dangerous factor that causes fatalities. However, the process of monitoring the scaffold is labor-intensive because it is conducted by the subjective observation of safety managers. To address this issue, we propose an automatic scaffold 3d reconstruction method using 3d point cloud data acquired using a robot dog. The method consists of three steps: 1) data acquisition of scaffold point clouds through a robot dog scanning system, 2) deep learning-based 3d semantic segmentation, and 3) automatic formation of a 3d CAd model. We created 15 robot dog datasets for training the segmentation model. The proposed method was tested at a different site where a scaffold with a representative structure was attached to the wall. The proposed method demonstrated an excellent performance of point cloud segmentation with a 90.84% F1 score.

关键词： Scaffold monitoring 3d point cloud data Robot dog Mobile laser scanning Transfer learning 3d semantic 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：