Filters








1,494 Hits in 5.7 sec

Semantics-Guided Moving Object Segmentation with 3D LiDAR [article]

Shuo Gu, Suling Yao, Jian Yang, Hui Kong
2022 arXiv   pre-print
In this paper, we propose a semantics-guided convolutional neural network for moving object segmentation. The network takes sequential LiDAR range images as inputs.  ...  Instead of segmenting the moving objects directly, the network conducts single-scan-based semantic segmentation and multiple-scan-based moving object segmentation in turn.  ...  The flowchart of the proposed semantics-guided moving object segmentation method from LiDAR point cloud.  ... 
arXiv:2205.03186v1 fatcat:wp5eli52abcl5ot62njejstpt4

Efficient Spatial-Temporal Information Fusion for LiDAR-Based 3D Moving Object Segmentation [article]

Jiadai Sun, Yuchao Dai, Xianjing Zhang, Jintao Xu, Rui Ai, Weihao Gu, Xieyuanli Chen
2022 arXiv   pre-print
How to effectively exploit the spatial-temporal information is a critical question for 3D LiDAR moving object segmentation (LiDAR-MOS).  ...  Accurate moving object segmentation is an essential task for autonomous driving.  ...  There are currently only a few datasets available for 3D LiDAR-based MOS, and the ratio of moving objects in the current Semantic-KITTI MOS dataset is relatively small.  ... 
arXiv:2207.02201v1 fatcat:t5eb7yasjrhfpgyrinxyjo6hiq

Deep Lidar CNN to Understand the Dynamics of Moving Vehicles [article]

Victor Vaquero, Alberto Sanfeliu, Francesc Moreno-Noguer
2018 arXiv   pre-print
These tasks include semantic information about vehicleness and a novel lidar-flow feature which combines standard image-based optical flow with lidar scans.  ...  In this paper we propose a novel solution to understand the dynamics of moving vehicles of the scene from only lidar information.  ...  detection, semantic segmentation or optical flow prediction [4] , [5] , [6] , [7] .  ... 
arXiv:1808.09526v2 fatcat:efqc6ofepfe4hoehnbfm56p56a

Teachers in concordance for pseudo-labeling of 3D sequential data [article]

Awet Haileslassie Gebrehiwot, Patrik Vacek, David Hurych, Karel Zimmermann, Patrick Perez, Tomáš Svoboda
2022 arXiv   pre-print
We show the performance of our method applied to multiple model architectures with tasks of 3D semantic segmentation and 3D object detection on two benchmark datasets.  ...  The output of multiple teachers is combined via a novel pseudo-label confidence-guided criterion. Our experimental evaluation focuses on the 3D point cloud domain in urban driving scenarios.  ...  A semisupervised learning to 3D semantic segmentation with guided point contrastive loss has been proposed in [12] .  ... 
arXiv:2207.06079v1 fatcat:xeeu6sthxnco5ishc33g5htmdi

Meta-RangeSeg: LiDAR Sequence Semantic Segmentation Using Multiple Feature Aggregation [article]

Song Wang, Jianke Zhu, Ruixiang Zhang
2022 arXiv   pre-print
We have conducted extensive experiments for performance evaluation on SemanticKITTI, which is the de-facto dataset for LiDAR semantic segmentation.  ...  To tackle these problems, we propose a novel approach to semantic segmentation for LiDAR sequences named Meta-RangeSeg, where a novel range residual image representation is introduced to capture the spatial-temporal  ...  It can be observed that our approach can effectively distinguish both static and moving objects with their semantic information. D.  ... 
arXiv:2202.13377v2 fatcat:ak5nngorhvhcvigyyoes56ovwi

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception [article]

Xinge Zhu, Hui Zhou, Tai Wang, Fangzhou Hong, Wei Li, Yuexin Ma, Hongsheng Li, Ruigang Yang, Dahua Lin
2021 arXiv   pre-print
State-of-the-art methods for driving-scene LiDAR-based perception (including point cloud semantic segmentation, panoptic segmentation and 3D detection, \etc) often project the point clouds to 2D space  ...  Furthermore, the proposed 3D framework also shows strong performance and good generalization on LiDAR panoptic segmentation and LiDAR 3D detection.  ...  Objective Function For LiDAR-based semantic segmentation task, the total objective of our method consists of two components, including voxel-wise loss and point-wise loss.  ... 
arXiv:2109.05441v1 fatcat:gmsqejbktjbkxiot5jvjal4xei

Fusion Based Holistic Road Scene Understanding [article]

Wenqi Huang, Xiaojin Gong
2014 arXiv   pre-print
Specifically, we first generate semantic object hypotheses by clustering 3D points, learning their prior appearance models, and using a deep learning method for reasoning their semantic categories.  ...  With this formulation, visual and range data are fused thoroughly, and moreover, the coupled segmentation and semantic labeling problem can be inferred via Graph Cuts.  ...  The model also integrates observed features of the pixels, together with the 3D lidar points and geometric contextual information to boost the accuracy of both object-level segmentation and semantic region  ... 
arXiv:1406.7525v1 fatcat:xnv5g2xojbgxpp253jookcjg6m

MASS: Multi-Attentional Semantic Segmentation of LiDAR Data for Dense Top-View Understanding [article]

Kunyu Peng, Juncong Fei, Kailun Yang, Alina Roitberg, Jiaming Zhang, Frank Bieder, Philipp Heidenreich, Christoph Stiller, Rainer Stiefelhagen
2022 arXiv   pre-print
While most previous works focus on sparse segmentation of the LiDAR input, dense output masks provide self-driving cars with almost complete environment information.  ...  ., through semantic segmentation of LiDAR sequences, which experienced a remarkable progress due to the release of large datasets such as SemanticKITTI and nuScenes-LidarSeg.  ...  Compared with GndNet [20] which predicts point-wise semantic segmentation category for each 3D LiDAR point, the top-view dense semantic segmentation map encodes higher-level semantic meanings especially  ... 
arXiv:2107.00346v2 fatcat:36a6n5znivch5n5g4ssffh7zqa

Semantic Mapping with Low-Density Point-Clouds for Service Robots in Indoor Environments

Carlos Medina Sánchez, Matteo Zella, Jesús Capitán, Pedro J. Marrón
2020 Applied Sciences  
LiDARs together with a set of practical segmentation methods for the detection of objects.  ...  Detecting and positioning objects as well as people in an accurate semantic map are, therefore, essential tasks that a robot needs to carry out.  ...  Thus, there are recent examples of methods for geometric and semantic 3D mapping with LiDARs [18] [19] [20] .  ... 
doi:10.3390/app10207154 fatcat:3ns6v6hwn5c2dl4onnwukasozm

Review on 3D Mapping and Segmentation

2020 International Journal of Engineering and Advanced Technology  
Mapping techniques include those using RGB images, RGBD images and LIDAR. Segmentation techniques include PointNet, PointNet++, 3D semantic and instance segmentation and joint instance segmentation.  ...  We concluded two key tasks for this purpose, which are 3D mapping and segmentation. This paper shows a comprehensive review of the different 3D mapping and segmentation methods.  ...  The method with the LIDAR input constructs a 3D model by processing high-density LIDAR data points.  ... 
doi:10.35940/ijeat.e1020.089620 fatcat:jjlyn3rqr5acllx7cahu3m5ok4

Semantic 3D Models from Real World Scene Recordings for Traffic Accident Simulation

Ludwig Mohr, Martin Öttl, Michael Haberl, Matthias Rüther, Horst Bischof
2018 Zenodo  
We present an enclosed pipeline generating 3D objects, their extents and relative positions as well as their semantic class from a combination of photogrammetric recordings and LiDAR (Light Detection And  ...  We propose a novel extension to traffic accident simulation by means of semantic 3D environment information allowing for a broader view by incorporating the entire close-by environment.  ...  yielding 3D objects, their extents and relative positions, as well as their semantic class, from a combination of photogrammetric recordings and LiDAR scans.  ... 
doi:10.5281/zenodo.1487620 fatcat:mniny6y4fvgybgm2d2mkjn4egi

A Survey of Simultaneous Localization and Mapping with an Envision in 6G Wireless Networks [article]

Baichuan Huang, Jun Zhao, Jingbin Liu
2020 arXiv   pre-print
The open question and forward thinking with an envision in 6G wireless networks end the paper.  ...  The paper makes an overview in SLAM including Lidar SLAM, visual SLAM, and their fusion.  ...  Squeezeseg: Convolutional neural nets with recurrent crf for real-time road-object segmentation from 3d lidar point cloud.  ... 
arXiv:1909.05214v4 fatcat:itnluvkewfd6fel7x65wdgig3e

LiDAR-Based Real-Time Panoptic Segmentation via Spatiotemporal Sequential Data Fusion

Weiqi Wang, Xiong You, Jian Yang, Mingzhan Su, Lantian Zhang, Zhenkai Yang, Yingcai Kuang
2022 Remote Sensing  
Subsequently, by improving the codec network, the multiscale features shared by semantic and instance branches were efficiently aggregated to achieve accurate panoptic segmentation for each LiDAR scan.  ...  An emerging research topic, panoptic segmentation, serves such a purpose by performing the tasks of semantic segmentation and instance segmentation in a unified framework.  ...  Compared with the semantic segmentation of a single scan, moving object semantic segmentation adds six categories: moving car, moving truck, moving other vehicle, moving person, moving bicyclist, and moving  ... 
doi:10.3390/rs14081775 fatcat:qe4mhwschjej3gib5vkgdc5kdy

Patch-Based Semantic Labeling of Road Scene Using Colorized Mobile LiDAR Point Clouds

Huan Luo, Cheng Wang, Chenglu Wen, Zhipeng Cai, Ziyi Chen, Hanyun Wang, Yongtao Yu, Jonathan Li
2016 IEEE transactions on intelligent transportation systems (Print)  
Index Terms-Semantic labeling, 3D-PMG, Markov random field, colorized mobile LiDAR point clouds.  ...  Then, to rectify the transferring errors caused by local patch similarities in different categories, contextual information among 3-D patches is exploited by combining 3D-PMG with Markov random fields.  ...  To alleviate the challenge of category imbalance, the prior information such as category label is introduced into the proposed algorithm to guide move searches.  ... 
doi:10.1109/tits.2015.2499196 fatcat:slu74a44njdjvaomtrqf7p5j6e

3D object-based classification for vehicle extraction from airborne LiDAR data by combining point shape information with spatial edge

Wei Yao, Stefan Hinz, Uwe Stilla
2010 2010 IAPR Workshop on Pattern Recognition in Remote Sensing  
Object extraction relies on the partitioning of raw ALS data into various segments approximating semantic entities followed by classification.  ...  A 3D segmentation method working directly on point cloud is used, which features the detection of local arbitrary modes and the globally optimized organization of segments concurrently.  ...  The moving object could generate motion artifacts in LiDAR data which are utilized to distinguish the movement and estimate the velocity.  ... 
doi:10.1109/prrs.2010.5742804 fatcat:n37zf6l4nffy3fnn2v5rtgjtdm
« Previous Showing results 1 — 15 out of 1,494 results