Paris-Lille-3D: A Point Cloud Dataset for Urban Scene Segmentation and Classification

Roynard, Xavier; Deschaud, Jean-Emmanuel; Goulette, François

doi:10.1109/cvprw.2018.00272

Cited by 36 publications

(15 citation statements)

References 9 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, simulations with a benchmark dataset are included to evaluate the proposed normal-vector and voxel-based ground filtering approaches. The selected benchmark is the Paris-Lille-3D dataset [19], which contains ten classes,…”

Section: Resultsmentioning

confidence: 99%

Real-time ground filtering algorithm of cloud points acquired using Terrestrial Laser Scanner (TLS)

Diaz,

Gallo,

Caceres

et al. 2021

Preprint

View full text Add to dashboard Cite

3D modeling based on point clouds requires ground-filtering algorithms that separate ground from non-ground objects. This study presents two ground filtering algorithms. The first one is based on normal vectors. It has two variants depending on the procedure to compute the k-nearest neighbors. The second algorithm is based on transforming the cloud points into a voxel structure. To evaluate them, the two algorithms are compared according to their execution time, effectiveness and efficiency. Results show that the ground filtering algorithm based on the voxel structure is faster in terms of execution time, effectiveness, and efficiency than the normal vector ground filtering.

show abstract

Section: Resultsmentioning

confidence: 99%

Real-time ground filtering algorithm of cloud points acquired using Terrestrial Laser Scanner (TLS)

Diaz,

Gallo,

Caceres

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Paris-Lille-3D. Paris-Lille-3D (Roynard et al, 2018) is a medium-size dataset that provides three aggregated point clouds. It is collected with a tilted rear-mounted Velodyne HDL-32E placed on a vehicle.…”

Section: Datasetsmentioning

confidence: 99%

“…We validate our approach on different scenarios getting state-of-the-art results. We use the Se-manticKitti (Behley et al, 2019) as the source domain and we adapt it to SemanticPoss (Pan et al, 2020), to Paris-Lille-3D (Roynard et al, 2018) and to a new collected and released dataset.…”

Section: Introductionmentioning

confidence: 99%

Domain Adaptation in LiDAR Semantic Segmentation by Aligning Class Distributions

Alonso¹,

Montesano²,

Murillo

2021

Proceedings of the 18th International Conference on Informatics in Control, Automation and Robotics

View full text Add to dashboard Cite

LiDAR semantic segmentation provides 3D semantic information about the environment, an essential cue for intelligent systems, such as autonomous vehicles, during their decision making processes. Unfortunately, the annotation process for this task is very expensive. To overcome this, it is key to find models that generalize well or adapt to additional domains where labeled data is limited. This work addresses the problem of unsupervised domain adaptation for LiDAR semantic segmentation models. We propose simple but effective strategies to reduce the domain shift by aligning the data distribution on the input space. Besides, we present a learningbased module to align the distribution of the semantic classes of the target domain to the source domain. Our approach achieves new state-of-the-art results on three different public datasets, which showcase adaptation to three different domains.

show abstract

“…Segmented and classified point clouds are used for vehicle localization (Maddern et al, 2015) and navigation (Chu et al, 2017). Two densely annotated point clouds of urban outdoor scenes are semantic3D.net (Hackel et al, 2017) and Paris-Lille-3D (Roynard et al, 2018). The former consists of dense point clouds of a wide range of outdoor scenes (churches, streets, railroad tracks, squares, villages, soccer fields and castles) whereas the latter is composed of point clouds of outdoor scenes in Paris and Lille with more than 50 classes.…”

Section: Introductionmentioning

confidence: 99%

CLOI: A Shape Classification Benchmark Dataset for Industrial Facilities

Agapaki¹,

Glyn-Davies²,

Mandoki³

et al. 2019

Computing in Civil Engineering 2019

View full text Add to dashboard Cite

Generation of digital models of existing industrial facilities is labor intensive and expensive. The use of state-of-the-art deep learning algorithms can assist to reduce the modelling time and cost. However large databases of labelled, laser-scanned industrial facilities do not exist to date, henceforth training of deep learning models is not possible. Our paper solves this problem by proposing a new benchmark dataset, which consists of five labelled industrial plants. The labelling schema that we followed for the generation of this dataset is based on the frequency of appearance of industrial object types. We labelled the ten most frequent industrial object shapes as identified in previous work. We present CLOI (Channels, L-shapes, circular sections, I-shapes): a richly annotated large-scale repository of shapes represented by labelled point clusters. CLOI has more than 140 million hand labelled points and serves as the foundation for researchers who are interested in automated modelling of industrial assets using deep learning algorithms.

show abstract

Paris-Lille-3D: A Point Cloud Dataset for Urban Scene Segmentation and Classification

Cited by 36 publications

References 9 publications

Real-time ground filtering algorithm of cloud points acquired using Terrestrial Laser Scanner (TLS)

Real-time ground filtering algorithm of cloud points acquired using Terrestrial Laser Scanner (TLS)

Domain Adaptation in LiDAR Semantic Segmentation by Aligning Class Distributions

CLOI: A Shape Classification Benchmark Dataset for Industrial Facilities

Contact Info

Product

Resources

About