A hierarchical methodology for urban facade parsing from TLS point clouds

Li, Zhuqiang; Zhang, Liqiang; Mathiopoulos, P. Takis; Liu, Fangyu; Zhang, Liang; Li, Shuaipeng; Líu, Hao

doi:10.1016/j.isprsjprs.2016.11.008

Cited by 42 publications

(23 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) Spin image feature Spin image [6,16] can express the shape features of the adjacent region for a point in three-dimensional space. Due to the strong robustness to occlusion and background interference and the insensitivity to rigid transformation of spin image feature, it is widely used in point clouds registration and three-dimensional objects recognition [6,9,16,19,33,[38][39][40]. Its specific extraction process is descried as follows.…”

Section: Multi-scale Single Point Features Extractionmentioning

confidence: 99%

“…Higher level features are mainly extracted by manifold learning [9,14], low-rank representation [15], sparse representation [6,16], and so on [17,18]. The most popular classifiers mainly include linear classifiers [19], random forests [20], AdaBoost [21], and SVM (support vector machine) [22]. For example, Mei et al [9] extracted color information, normal vector, spin image, and elevation features of each point using nearest neighbor points selected by radius.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Point Set Multi-Level Aggregation Feature Extraction Based on Multi-Scale Max Pooling and LDA for Point Cloud Classification

Tong

Zhang

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

Accurate and effective classification of lidar point clouds with discriminative features expression is a challenging task for scene understanding. In order to improve the accuracy and the robustness of point cloud classification based on single point features, we propose a novel point set multi-level aggregation features extraction and fusion method based on multi-scale max pooling and latent Dirichlet allocation (LDA). To this end, in the hierarchical point set feature extraction, point sets of different levels and sizes are first adaptively generated through multi-level clustering. Then, more effective sparse representation is implemented by locality-constrained linear coding (LLC) based on single point features, which contributes to the extraction of discriminative individual point set features. Next, the local point set features are extracted by combining the max pooling method and the multi-scale pyramid structure constructed by the point’s coordinates within each point set. The global and the local features of the point sets are effectively expressed by the fusion of multi-scale max pooling features and global features constructed by the point set LLC-LDA model. The point clouds are classified by using the point set multi-level aggregation features. Our experiments on two scenes of airborne laser scanning (ALS) point clouds—a mobile laser scanning (MLS) scene point cloud and a terrestrial laser scanning (TLS) scene point cloud—demonstrate the effectiveness of the proposed point set multi-level aggregation features for point cloud classification, and the proposed method outperforms other related and compared algorithms.

show abstract

Section: Multi-scale Single Point Features Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Point Set Multi-Level Aggregation Feature Extraction Based on Multi-Scale Max Pooling and LDA for Point Cloud Classification

Tong

Zhang

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…More recently, Li et al. () introduced a novel hierarchical approach to semantically segment urban environmental façade objects. In their approach, the input point cloud was first converted into individual plane objects.…”

Section: Related Workmentioning

confidence: 99%

Automated Analysis of Mobile LiDAR Data for Component‐Level Damage Assessment of Building Structures during Large Coastal Storm Events

Zhou

Gong

2018

Computer aided Civil Eng

View full text Add to dashboard Cite

Rapid assessment of building damages due to natural disasters is a critical element in disaster management. Although airborne-based remote sensing techniques have been successfully applied in many postdisaster scenarios, automated building componentlevel damage assessment with terrestrial/mobile LiDAR data is still challenging to achieve due to lack of reliable segmentation methods for damaged buildings. In this research, a novel building segmentation and damage detection approach is proposed to realize automated component-level damage assessment for major building envelop elements including wall, roof, balcony, column, and handrail. The proposed approach first conducts semantic segmentation of building point cloud data using a rule-based approach. The detected building components are then evaluated to determine if the components are damaged. The authors applied this method on a mobile LiDAR data set collected after Hurricane Sandy. The results demonstrate that the approach is capable of achieving 96% and 86% parsing accuracy for wall façades and roof facets, and obtain 82% and 78% of detection accuracy for damaged walls and roof facets. C 2018 Computer-Aided Civil and Infrastructure Engineering.

show abstract

“…It is not an overstatement to say that there is no complete, end-to-end automated process for In one hand, there is prior work on modelling the exteriors of buildings, but mostly has been focused on the reconstruction of planar façades with, at best, openings like windows and doors. Authors like Wan (2012), Wang (2015) and Li (2017) proposed algorithms to model building exteriors recognizing walls as planes. These resulting planes are grouped together into a single building volume.…”

Section: Introductionmentioning

confidence: 99%

Towards the Automatic 3d Parametrization of Non-Planar Surfaces From Point Clouds in Hbim Applications

Román¹,

Lerones²,

Llamas³

et al. 2019

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

3D laser scanning and photogrammetric 3D reconstruction generate point clouds from which the geometry of BIM models can be created. However, a few methods do this automatically for concrete architectural elements, but in no case for the entirety of heritage assets. A novel procedure for the automatic recognition and parametrization of non-planar surfaces of heritage immovable assets is presented using point clouds as raw input data. The methodology is able to detect the most relevant architectural features in a point cloud and their interdependences through the analysis of the intersections of related elements. The non-planar surfaces detected, mainly cylinders, are studied in relation to the neighbouring planar surfaces present in the cloud so that the boundaries of both the planar and the non-planar surfaces are accurately defined. The procedure is applied to the emblematic Castle of Torrelobatón, located in Valladolid (Spain) to allow the cataloguing of required elements, as illustrative example of the European defensive architecture from the Middle age to the Renaissance period. Results and conclusions are reported to evaluate the performance of this approach.

show abstract

A hierarchical methodology for urban facade parsing from TLS point clouds

Cited by 42 publications

References 23 publications

Point Set Multi-Level Aggregation Feature Extraction Based on Multi-Scale Max Pooling and LDA for Point Cloud Classification

Point Set Multi-Level Aggregation Feature Extraction Based on Multi-Scale Max Pooling and LDA for Point Cloud Classification

Automated Analysis of Mobile LiDAR Data for Component‐Level Damage Assessment of Building Structures during Large Coastal Storm Events

Towards the Automatic 3d Parametrization of Non-Planar Surfaces From Point Clouds in Hbim Applications

Contact Info

Product

Resources

About