Multi-scale EM-ICP: A Fast and Robust Approach for Surface Registration

By Sébastien Granger, Xavier Pennec. European Conference on Computer Vision, 2002

505 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computational Mechanics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c76319e8f9e61e058c48f877f489c6157c1fbe9b

How researchers see this paper

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

In Aerospace Engineering

In Computational Mechanics

In Computer Vision and Pattern Recognition

Citations in the corpus

Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration (2016, methodology): “Granger and Pennec [33] proposed an algo-rithm named Multi-scale EM-ICP where an annealing scheme on GMM variance was also used.”
Fast LIDAR localization using multiresolution Gaussian mixture maps (2015, methodology): “We employ techniques presented by Olson [22] to formulate the multiresolution search as a branch-and-bound problem that can guarantee global optimality over our search space.”
Shape Registration with Spherical Cross Correlation (2008, unknown): “Granger [5] introduced the EM-ICP method.”
A robust algorithm for point set registration using mixture of Gaussians (2005, background): “Also, a naive implementation of ICP is know to be prone to outliers which prompted several more robust variations [1, 2].”
Fast Global Registration (2016, background): “Other works explored the utility of relaxed assignments [14, 24, 32], distance field representations [6], and mixture models [21, 41] for increasing the robustness of local registration.”
A Correlation-Based Approach to Robust Point Set Registration (2004, methodology): “KC registration is mathematically most related to the EM-ICP algorithm [6] and the SoftAssignment algorithm [14], which are also multiply-linked ICP.”
Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “Numerous studies, such as EM-ICP (Granger and Pennec 2002), LM-ICP (Fitzgibbon 2003), and TriICP (Chetverikov et al.”
Deformable Medical Image Registration: A Survey (2013, methodology): “Pennec et al. [23] dropped the linearity assumption by modeling the deformation process through the St Venant-Kirchoff elasticity energy that extends the linear elastic model to the nonlinear regime.”
TEASER: Fast and Certifiable Point Cloud Registration (2020, background): “In general, the literature is divided between heuristics, that are fast but brittle, and global methods that are guaranteed to be robust, but run in worst-case exponential time (e.g., branch-&-bound methods, such as Go-ICP [20]).”