Title: GCSAC: geometrical constraint sample consensus for primitive shapes estimation in 3D point cloud

Authors: Van-Hung Le; Hai Vu; Thuy Thi Nguyen; Thi-Lan Le; Thanh-Hai Tran

Addresses: International Research Institute MICA, Hanoi University of Science and Technology, 1 Dai Co Viet Street, Hai Ba Trung, Hanoi, Vietnam; Tan Trao University, QL2, Trung Môn, Yên Sơn, Tuyên Quang, Vietnam ' International Research Institute MICA, Hanoi University of Science and Technology, 1 Dai Co Viet Street, Hai Ba Trung, Hanoi, Vietnam ' Faculty of Information Technology, VietNam National University Agriculture, Trâu Quỳ, Gia Lâm, Hanoi, Vietnam ' International Research Institute MICA, Hanoi University of Science and Technology, 1 Dai Co Viet Street, Hai Ba Trung, Hanoi, Vietnam ' International Research Institute MICA, Hanoi University of Science and Technology, 1 Dai Co Viet Street, Hai Ba Trung, Hanoi, Vietnam

Abstract: Estimating parameters of a primitive shape from a point cloud data is a challenging problem due to the data containing noises and computational time demand. In this paper, we present a new robust estimator (named GCSAC, geometrical constraint sample consensus) aimed at solving such issues. The proposed algorithm takes into account geometrical constraints to construct qualified samples for the estimation. Instead of randomly drawing minimal subset of sample, explicit geometrical properties of the interested primitive shapes (e.g., cylinder, sphere and cone) are used to drive the sampling procedures. Based on the collected samples, model estimation and verification procedures of the robust estimator are deployed in GCSAC. Extensive experiments are conducted on synthesised and real datasets. Comparing with the common algorithms of RANSAC family, GCSAC outperforms in term of both the precision of the estimated model and computational time. The implementations of GCSAC and the datasets are made publicly available.

Keywords: robust estimator; primitive shape estimation; random sample consensus; RANSAC and RANSAC variations; quality of samples; point cloud data.

DOI: 10.1504/IJCVR.2019.101539

International Journal of Computational Vision and Robotics, 2019 Vol.9 No.4, pp.387 - 411

Received: 30 Dec 2017
Accepted: 28 Aug 2018
Published online: 12 Aug 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article