Efficient RANSAC for Point‐Cloud Shape Detection

By Ruwen Schnabel, Roland Wahl, Reinhard Klein. 2007

2,286 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Geology, 3D Vision & Pose, Algorithms & Complexity

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Geology

In 3D Vision & Pose

In Algorithms & Complexity

Citations in the corpus

3D Instance Segmentation of MVS Buildings (2021, methodology): “Then, we reduce the dimension of these boundary edges to the 2D horizontal plane by discarding their height and further simplify them through a RANSAC process [47].”
On-machine and in-process surface metrology for precision manufacturing (2019, neutral): “The best fit of this discrete model to the set of measured points needs segmentation [202], registration of 3D point clouds [237] and alignment methods [37].”
Incremental and Enhanced Scanline-Based Segmentation Method for Surface Reconstruction of Sparse LiDAR Data (2016, methodology): “For each combined segment, we use RANSAC to detect whether a segment is a planar shape and estimate the parameters of the plane if it is.”
SnapNet: 3D point cloud semantic labeling with 2D deep segmentation networks (2017, background): “[ 19] proposes a generic RANSAC algorithm for geometric shape extraction in 3D point clouds.”
Creating Large-Scale City Models from 3D-Point Clouds: A Robust Approach with Hybrid Representation (2012, neutral): “As the classi-ﬁcation proposed in Section 2 rejects outliers from the building point set, the use of Ransac-based algorithms, which are more computationally expensive for similar problems (Schnabel et al., 2007; Toshev et al., 2010), is not required.”
Deep Learning Advances in Computer Vision with 3D Data (2017, methodology): “Furthermore, clustering algorithms [Douillard et al. 2011], 3D Hough Trans-form [Borrmann et al. 2011], RANSAC [Schnabel et al. 2007], and probabilistic models, such as Markov Random Fields (MRFs) and Conditional Random Fields (CRFs), have also been employed (e.g., Valentin et al. [2013]).”
An update on automatic 3D building reconstruction (2010, positive): “Common shapes like planes, spheres, cylinders, cones and tori are iteratively detected in the DSM by a probabilistic RANSACbased algorithm as described in Schnabel et al. (2007).”
State of research in automatic as-built modelling (2015, positive): “propose a RANSAC version based on a new sampling strategy together with an early termination scheme that provides a significant speedup [100].”
3D Semantic Parsing of Large-Scale Indoor Spaces (2016, methodology): “Ours RANSAC 2D Hough Building final over-segm based [27] based [38]”
Reconstructing building mass models from UAV images (2016, methodology): “After the 2D filtering preprocess, we generate a set of plane hypotheses from the depth map using RANSAC [34].”
A Survey of Surface Reconstruction from Point Clouds (2017, methodology): “In cases c 2016Authorversion where we are concerned with watertight reconstruction of individual objects, the detection of primitives [SWK07] can subsequently be used for primitive extrapolation for reconstruction when faced with large amounts of missing data [SDK09].”
Towards 3D Point cloud based object maps for household environments (2008, methodology): “An automatic algorithm to detect basic shapes in unorganized point clouds is presented in [34].”