A Survey of Simple Geometric Primitives Detection Methods for Captured 3D Data

Comparison webapp for the presented algorithms, details in Section 5.5 of the paper.

Usage: click on the column title to reorder the list of method based on this criterion.

References are listed at the end of this page.



---Not applicable / Information not available in paper
○Moderate / Optional
●Good / Required
RGB image
Depth image
RGB-D image
Point Cloud
△Mesh
□Voxel
⁎

Sequence or Set of images

Year
Author
Name
Planes
Boxes
Cuboids
Spheres
Cylinders
Cones
Other Primitives
Context
Input
Inner Data
Detection Category
Data Fidelity
Abstraction Level
Real-time
Scalability
User Assistance
Learning Phase
Intuitive Tuning
Temporal Consistency
Semantics
Needs Extra Information
Provides Meta Data
Robustness to Noise
Robustness to Incomplete Data
Robustness to Outliers
Code Available
[1]2016OchmannWalls Layout✓building interiorsRANSAC23●●---○○○●
[2]2016OesauRegular Planar Modeling✓building interiorsprimitive growing33●●---○○○
[3]2015AlehdaghiParallel RANSAC✓indoor sceneprimitive growing22●●●●
[4]2015BagautdinovOccupancy Maps✓indoor scene local statistics13------●●●
[5]2015ElghorPlanar RGB-D SLAM✓indoor scene ⁎RANSAC22●●●●●●●
[6]2015KaessQuaternion Representation✓indoor scene ⁎RANSAC22●●●●●○●
[7]2015KangBayes SAmple Consensus✓✓✓anyRANSAC32●○---○●●
[8]2015KhanBoxes around Objects✓indoor sceneprimitive growing13---●●---●●●○●
[9]2015Limberger3D Kernel Hough Transform✓indoor sceneparameter space33●●●---●●●●
[10]2015Martinovic3D All The Way✓urban buildings ⁎segmentation + fitting23●●○○●●●○○●
[11]2015MonszpartRAPter✓building interiorsprimitive growing23●●---●●●●
[12]2015VerdieLevel of Detail✓✓✓urban buildings△△segmentation + fitting33●○---●○●○●
[13]2015ZhangLabeled KinectFusion✓indoor scene ⁎□primitive growing32●●○○●●●○
[14]2015ZhouGeneralized Cylinder✓individual objects△△primitive growing32---○---○○●
[15]2014ChenSemantic Modeling✓indoor scene ⁎RANSAC32---●●---●●○●○
[16]2014FengAgglomerative Clustering✓indoor sceneprimitive growing32●●●●●●●
[17]2014HulikPlanar Hough Transform✓indoor scene ⁎parameter space23○○○●●○○
[18]2014MattauschCluttered Indoor Scans✓building interiorsprimitive growing32●○---○○●●○●
[19]2014OchmannHierarchical Building Descriptions✓✓building interiorsRANSAC32●●○---○○●●●○
[20]2014SalasDense Planar SLAM✓indoor scene ⁎primitive growing22●●○●●●○●
[21]2013ArikanO-Snap✓urban buildingsRANSAC33●○○---●●●
[22]2013Chen3-Sweep✓✓individual objectsuser-assisted32●●●---●
[23]2013JiangFitting Cuboids✓indoor sceneprimitive growing12------●○○●●
[24]2013KimLearning Object Templates✓individual objects△ ⁎segmentation + fitting12---○○---○○●○●●
[25]2013LafargePoint Set Structuring✓urban buildingsRANSAC22●●---●●●
[26]2013LinSemantic Learning✓✓✓residential sceneprimitive growing33●●●●---●○●●●
[27]2013ShtofGeosemantic Snapping✓✓✓✓individual objectsuser-assisted33●●●---●●●●
[28]2013TaguchiPoint-plane SLAM✓indoor scene ⁎RANSAC21●●●●●●
[29]2013ThierySphere Meshes✓✓individual objects△△primitive growing23○●●---○●
[30]2013Trevor2D-structured Point Cloud✓indoor sceneprimitive growing22●○○○●○
[31]2012BiswasIndoor Robot Navigation✓building interiors ⁎RANSAC11●●●●●●●
[32]2012CarrMonocular Occupancy Maps✓✓anylocal statistics13●○○○●●●●
[33]2012GoronObjects On Table✓✓household scenesegmentation + fitting33---●---●●●
[34]2012LafargeHybrid City Representation✓✓✓✓urban buildingssegmentation + fitting32●○---●●●
[35]2012LeeIndoor Plane Mapping✓indoor scene ⁎primitive growing11●●●●●
[36]2012ShaoInteractive Semantic Modeling✓indoor scene ⁎primitive growing22●●○●●●●
[37]2012SilbermanObject Support✓indoor sceneRANSAC11---●●●●
[38]2012TrevorPlanar Surface SLAM✓indoor scene ⁎RANSAC22---○●○○●
[39]2012WoodfordContracting Segments✓✓✓anyprimitive growing21○○---○●
[40]2012XiaoOutdoor Planar SLAM✓urban buildings ⁎primitive growing21---●●●●●
[41]2012YanQuadric Surface Fitting✓✓✓✓individual objects△△automatic clustering31○●---●○
[42]2011BeniereCAD Model Recovery✓✓✓✓individual objects△△primitive growing32○---●
[43]2011BiswasPlane Filtering✓indoor scene ⁎RANSAC12●●●●●
[44]2011ChenAlgebraic Templates✓✓✓urban buildingsRANSAC33○●---●●●
[45]2011HolzCluster Normal Space✓indoor sceneparameter space12●○---○●○●
[46]2011LiGlobFit✓✓✓✓individual objectsRANSAC33○---●●●●●
[47]2011ShenFacade Partitioning✓urban buildingsRANSAC23○●---○●●●
[48]2011XiaoPlane Detection for SLAM✓indoor scene ⁎primitive growing22---●○●
[49]2010AtteneHierarchical Modeling✓✓✓✓✓individual objectsprimitive growing32○○○---●○●○●
[50]2009FurukawaManhattan World Stereo✓urban buildings ⁎parameter space22○○○○
[51]2009FurukawaVolumetric Integration✓indoor scene ⁎□parameter space21○●○●●○○
[52]2009RusuHybrid Object Model✓✓✓✓household scenesegmentation + fitting32---○---○○●
[53]2008ChenArchitectural Modeling✓urban buildingsparameter space23○●---●○
[54]2007RabbaniPrimitive-based registration✓✓✓✓indoor sceneparameter space22---○●○○●
[55]2007Rusu3D Object Maps✓indoor scene ⁎RANSAC21---●○○●●●
[56]2007SchnabelFast RANSAC✓✓✓✓✓individual objectsRANSAC32●○○---●●●●
[57]2006AtteneHierarchical Segmentation✓✓✓individual objects△△primitive growing22○○●---●○●●
[58]2006LalondeOutdoor Robot Navigation✓✓✓natural scenesegmentation + fitting23●●●●●●●
[59]2006WeingartenPatch RANSAC✓indoor scene ⁎△RANSAC11○●●●●
[60]2005RabbaniCylindric Hough Transform✓indoor sceneparameter space32---○●---●○●●
[61]2005SimariEllipsoidal Modeling✓organic shapes△△automatic clustering22---○●---●○
[62]2005WuHybrid VSA✓✓✓individual objects△△automatic clustering22○●---●○
[63]2004CohenVSA✓individual objects△△automatic clustering21●●---●○
[64]2004GelfandLocal Slippage Analysis✓✓✓✓individual objectsprimitive growing22------○---○○
[65]2004SchusterTensor Voting✓urban buildingsprimitive growing12------●●
[66]2004WangAssisted Model-Image Fitting✓✓✓✓urban buildingsuser-assisted12●●●○●●
[67]2003DecoretBillboard Clouds✓individual objects△△parameter space21●---○○●
[68]2002WangHough-based Reconstruction✓indoor scene ⁎parameter space12●○○---●●●
[69]2001MarshallGrow and Merge✓✓✓individual objectsprimitive growing23---○---○●
[70]2001VosselmanHough-based House Modeling✓residential sceneparameter space22○---○●○
[71]1999WhitakerPlane-based Registration✓indoor scene ⁎segmentation + fitting12●●●
[72]1998LukacsLeast-Squares Fitting✓✓✓✓individual objectsprimitive growing22------○


Available source code

(links accessed July 03, 2018)


MethodLanguage
[9]3D Kernel Hough TransformC++link
[11]RAPterC++link
[14]Generalized CylinderC++link
[16]Agglomerative ClusteringC++link
[18]Cluttered Indoor ScansC++link
[23]Fitting CuboidsMatlablink
[24]Learning Object TemplatesC++link
[27]Geosemantic SnappingC#link
[29]Sphere MeshesC++link
[31]Indoor Robot NavigationC++link
[37]Object SupportMatlablink
[43]Plane FilteringC++link
[46]GlobFitC++link
[48]Plane Detection for SLAMC++link
[56]Fast RANSAC (CGAL)C++link
[56]Fast RANSAC (original)C++link
[57]Hierarchical SegmentationC++link
[67]Billboard CloudsC++link
(PCL)Variants of RANSACC++link


Available datasets

(links accessed July 03, 2018)


MethodType of Data
[9]3D Kernel Hough Transform3D point cloudslink
[11]RAPterPoint cloud + Detected Primitiveslink
[14]Generalized Cylinder3D models (OBJ) + Parameters + Detected Primitiveslink
[15]Semantic ModelingRGB-D images + Generated Modellink
[18]Cluttered Indoor ScansPoint cloud + Normalslink
[24]Learning Object Templates3D models + Ground truthlink
SUN3DSequence of RGB-D images + Camera poses + Segmentationlink
[36]Interactive Semantic ModelingRGB-D images + Matched 3D modelslink
[37]NYU Depth Dataset V2Sequence of RGB-D images + Segmentationlink
[46]GlobFitPoint cloud from individual objects + Detected primitiveslink
[56]Fast RANSACPoint cloud from individual objects + Detected primitiveslink


Visual Compendium

(click on image to show article page, links accessed July 03, 2018)


                                                                                                                                                                                                                                                                                               


References

(links accessed July 03, 2018)


[1]Ochmann, Sebastian and Vock, Richard and Wessel, Raoul and Klein, ReinhardAutomatic Reconstruction of Parametric Building Models from Indoor Point Cloudslink
[2]Oesau, Sven and Lafarge, Florent and Alliez, PierrePlanar shape detection and regularization in tandemlink
[3]Alehdaghi, Mahdi and Esfahani, Mahdi Abolfazli and Harati, AhadParallel RANSAC: Speeding up plane extraction in RGBD image sequences using GPUlink
[4]Bagautdinov, Timur and Fleuret, Francois and Fua, PascalProbability Occupancy Maps for Occluded Depth Imageslink
[5]Elghor, Hakim Elchaoui and Roussel, David and Ababsa, Fakhreddine and Bouyakhf, El HoussinePlanes Detection for Robust Localization and Mapping in RGB-D SLAM systemslink
[6]M. KaessSimultaneous Localization and Mapping with Infinite Planeslink
[7]Kang, Zhizhong and Li, ZhenPrimitive fitting based on the efficient multiBaySAC algorithmlink
[8]Khan, Salman H. and He, Xuming and Bennamoun, Mohammed and Sohel, Ferdous and Togneri, RobertoSeparating Objects and Clutter in Indoor Sceneslink
[9]Limberger, Frederico A and Oliveira, Manuel MReal-time detection of planar regions in unorganized point cloudslink
[10]Martinovic, Andelo and Knopp, Jan and Riemenschneider, Hayko and Van Gool, Luc3D All The Way: Semantic Segmentation of Urban Scenes From Start to End in 3Dlink
[11]Aron Monszpart and Nicolas Mellado and Gabriel Brostow and Niloy MitraRAPter: Rebuilding Man-made Scenes with Regular Arrangements of Planeslink
[12]Verdie, Yannick and Lafarge, Florent and Alliez, PierreLOD Generation for urban sceneslink
[13]Zhang, Yizhong and Xu, Weiwei and Tong, Yiying and Zhou, KunOnline structure analysis for real-time indoor scene reconstructionlink
[14]Zhou, Yang and Yin, Kangxue and Huang, Hui and Zhang, Hao and Gong, Minglun and Cohen-Or, DanielGeneralized Cylinder Decompositionlink
[15]Chen, Kang and Lai, Yu-Kun and Wu, Yu-Xin and Martin, Ralph and Hu, Shi-MinAutomatic Semantic Modeling of Indoor Scenes from Low-quality RGB-D Data Using Contextual Informationlink
[16]Feng, Chen and Taguchi, Yasuhiro and Kamat, Vineet RFast plane extraction in organized point clouds using agglomerative hierarchical clusteringlink
[17]Hulik, Rostislav and Spanel, Michal and Smrz, Pavel and Materna, ZdenekContinuous plane detection in point-cloud data based on 3D Hough Transformlink
[18]Oliver Mattausch and Daniele Panozzo and Claudio Mura and Olga Sorkine-Hornung and Renato PajarolaObject Detection and Classification from Large-Scale Cluttered Indoor Scanslink
[19]Ochmann, Sebastian and Vock, Richard and Wessel, Raoul and Klein, ReinhardTowards the Extraction of Hierarchical Building Descriptions from 3D Indoor Scanslink
[20]Salas-Moreno, Renato F and Glocken, Ben and Kelly, Paul HJ and Davison, Andrew JDense Planar SLAMlink
[21]Murat Arikan and Michael Schwarzler and Simon Flory and Michael Wimmer and Stefan MaierhoferO-Snap: Optimization-Based Snapping for Modeling Architecturelink
[22]Chen, Tao and Zhu, Zhe and Shamir, Ariel and Hu, Shi-Min and Cohen-Or, Daniel3-sweep: Extracting editable objects from a single photolink
[23]Jiang, Hao and Xiao, JianxiongA linear approach to matching cuboids in RGBD imageslink
[24]Vladimir G. Kim and Li, Wilmot and Mitra, Niloy J. and Chaudhuri, Siddhartha and DiVerdi, Stephen and Funkhouser, ThomasLearning Part-based Templates from Large Collections of 3D Shapeslink
[25]Lafarge, Florent and Alliez, PierreSurface reconstruction through point set structuringlink
[26]Lin, Hui and Gao, Jizhou and Zhou, Yu and Lu, Guiliang and Ye, Mao and Zhang, Chenxi and Liu, Ligang and Yang, RuigangSemantic Decomposition and Reconstruction of Residential Scenes from LiDAR Datalink
[27]Shtof, Alex and Agathos, Alexander and Gingold, Yotam and Shamir, Ariel and Cohen-Or, DanielGeosemantic Snapping for Sketch-Based Modelinglink
[28]Taguchi, Yasuhiro and Jian, Yong-Dian and Ramalingam, Srikumar and Feng, ChenPoint-plane SLAM for hand-held 3D sensorslink
[29]Jean-Marc Thiery and Emilie Guy and Tamy BoubekeurSphere-Meshes: Shape Approximation using Spherical Quadric Error Metricslink
[30]Trevor, Alexander JB and Gedikli, Suat and Rusu, Radu B and Christensen, Henrik IEfficient organized point cloud segmentation with connected componentslink
[31]Biswas, Joydeep and Veloso, ManuelaDepth camera based indoor mobile robot localization and navigationlink
[32]Carr, Peter and Sheikh, Yaser and Matthews, IainMonocular object detection using 3d geometric primitiveslink
[33]Goron, Lucian Cosmin and Marton, Zoltan-Csaba and Lazea, Gheorghe and Beetz, MichaelRobustly segmenting cylindrical and box-like objects in cluttered scenes using depth cameraslink
[34]Lafarge, Florent and Mallet, ClementCreating large-scale city models from 3D-point clouds: a robust approach with hybrid representationlink
[35]Lee, Tae-kyeong and Lim, Seungwook and Lee, Seongsoo and An, Shounan and Oh, Se-youngIndoor mapping using planes extracted from noisy RGB-D sensorslink
[36]Shao, Tianjia and Xu, Weiwei and Zhou, Kun and Wang, Jingdong and Li, Dongping and Guo, BainingAn interactive approach to semantic modeling of indoor scenes with an RGBD cameralink
[37]Nathan Silberman and Derek Hoiem and Pushmeet Kohli and Rob FergusIndoor Segmentation and Support Inference from RGBD Imageslink
[38]Trevor, Alexander JB and Rogers III, John G and Christensen, Henrik IPlanar surface SLAM with 3D and 2D sensorslink
[39]Woodford, Oliver J and Pham, Minh-Tri and Maki, Atsuto and Gherardi, Riccardo and Perbet, Frank and Stenger, BjornContraction moves for geometric model fittinglink
[40]Xiao, Junhao and Adler, Bejamin and Zhang, Houxiang3D point cloud registration based on planar surfaceslink
[41]Yan, Dong-Ming and Wang, Wenping and Liu, Yang and Yang, ZhouwangVariational mesh segmentation via quadric surface fittinglink
[42]Beniere, Roseline and Subsol, Gerard and Gesquiere, Gilles and Le Breton, Franccois and Puech, WilliamRecovering primitives in 3D cad mesheslink
[43]Biswas, Joydeep and Veloso, ManuelaFast sampling plane filtering, polygon construction and merging from depth imageslink
[44]Chen, Jyun-Yuan and Lai, Hung-Jui and Lin, Chao-HungPoint cloud modeling using algebraic templatelink
[45]Holz, Dirk and Holzer, Stefan and Rusu, Radu Bogdan and Behnke, SvenReal-time plane segmentation using RGB-D cameraslink
[46]Yangyan Li and Xiaokun Wu and Yiorgos Chrysanthou and Andrei Sharf and Daniel Cohen-Or and Niloy J. MitraGlobFit: Consistently Fitting Primitives by Discovering Global Relationslink
[47]Shen, Chao-Hui and Huang, Shi-Sheng and Fu, Hongbo and Hu, Shi-MinAdaptive Partitioning of Urban Facadeslink
[48]Xiao, Junhao and Zhang, Jianhua and Zhang, Jianwei and Zhang, Houxiang and Hildre, Hans PetterFast plane detection for SLAM from noisy range images in both structured and unstructured environmentslink
[49]Attene, Marco and Patane, GiuseppeHierarchical Structure Recovery of Point-Sampled Surfaceslink
[50]Furukawa, Yudai and Curless, Brian and Seitz, Steven M and Szeliski, RichardManhattan-world stereolink
[51]Furukawa, Yasutaka and Curless, Brian and Seitz, Steven M and Szeliski, RichardReconstructing building interiors from imageslink
[52]Rusu, Radu Bogdan and Blodow, Nico and Marton, Zoltan Csaba and Beetz, MichaelClose-range scene segmentation and reconstruction of 3D point cloud maps for mobile manipulation in domestic environmentslink
[53]Chen, Jie and Chen, BaoquanArchitectural modeling from sparsely scanned range datalink
[54]Rabbani, Tahir and Dijkman, Sander and van den Heuvel, Frank and Vosselman, GeorgeAn integrated approach for modelling and global registration of point cloudslink
[55]Rusu, Radu Bogdan and Blodow, Nico and Marton, Zoltan and Soos, Alina and Beetz, MichaelTowards 3D object maps for autonomous household robotslink
[56]Schnabel, Ruwen and Wahl, Roland and Klein, ReinhardEfficient RANSAC for Point-Cloud Shape Detectionlink
[57]Attene, Marco and Falcidieno, Bianca and Spagnuolo, MichelaHierarchical mesh segmentation based on fitting primitiveslink
[58]Jean-Francois Lalonde and Nicolas Vandapel and Daniel Huber and Martial Hebert Natural terrain classification using three-dimensional ladar data for ground robot mobilitylink
[59]Weingarten, Jan and Siegwart, Roland3D SLAM using planar segmentslink
[60]Rabbani, Tahir and Van Den Heuvel, FrankEfficient hough transform for automatic detection of cylinders in point cloudslink
[61]Simari, Patricio D and Singh, KaranExtraction and remeshing of ellipsoidal representations from mesh datalink
[62]Wu, Jianhua and Kobbelt, LeifStructure recovery via hybrid variational surface approximationlink
[63]Cohen-Steiner, David and Alliez, Pierre and Desbrun, MathieuVariational shape approximationlink
[64]Gelfand, Natasha and Guibas, Leonidas JShape segmentation using local slippage analysislink
[65]Schuster, Hanns-FSegmentation of lidar data using the tensor voting frameworklink
[66]Wang, SD and Tseng, YHSemi-automated CSG model-based building extraction from photogrammetric imageslink
[67]Decoret, Xavier and Durand, Fredo and Sillion, Franccois X and Dorsey, JulieBillboard clouds for extreme model simplificationlink
[68]Wang, Jianning and Oliveira, Manuel MImproved scene reconstruction from range imageslink
[69]Marshall, David and Lukacs, Gabor and Martin, RalphRobust segmentation of primitives from range data in the presence of geometric degeneracylink
[70]Vosselman, George and Dijkman, Sander3D building model reconstruction from point clouds and ground planslink
[71]Whitaker, Ross T and Gregor, Jens and Chen, PFIndoor scene reconstruction from sets of noisy range imagelink
[72]Lukacs, Gabor and Martin, Ralph and Marshall, DaveFaithful least-squares fitting of spheres, cylinders, cones and tori for reliable segmentationlink