ICP Registration Using Invariant Features

By G. Sharp, S. W. Lee, D. Wehe. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002

686 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/31fe83670bfb78b7e895f5080f53c63ac33417e3

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 Computer Vision and Pattern Recognition

Citations in the corpus

SHREC 2007 3 D Shape Retrieval Contest (2006, methodology): “On the basis of these observations, we built our benchmark, shown in Figure 1.”
Registration of 3D Point Clouds and Meshes: A Survey from Rigid to Nonrigid (2013, background): “These features include spin images [52], mean curvature [56], Gaussian curvature [53], moments and spherical harmonics [54], integral descriptors [46], [48], [55], fast Fourier transform (FFT) and discrete cosine transform (DCT) coefficients of distributions of normals [17], cluster signatures of si”
Towards 3D lidar point cloud registration improvement using optimal neighborhood knowledge (2013, background): “The ‘‘reliability’’ may be evaluated according to planar criteria or with scalespace analysis (Sharp et al., 2002; Ho et al., 2009).”
Least squares 3D surface and curve matching (2005, methodology): “In Sharp et al. (2002) a conceptually similar method called Iterative Closest Points using Invariant Features (ICPIF) has been introduced.”
The dual-bootstrap iterative closest point algorithm with application to retinal image registration (2003, methodology): “By enriching the feature set, matching can be improved, both prior to ICP and during ICP [44, 65 , 74].”
Aligning point cloud views using persistent feature histograms (2008, methodology): “Since input datasets do not always contain complete pointto-point correspondences, due to the fact that they might be only partially overlapping for example, a lot of efforts have been made into the area of feature selection [3], as well as including extra information such as colors [4], intensity v”
A review of recent range image registration methods with accuracy evaluation (2007, methodology): “[33] proposed the ICP using invariant features ICPIF.”
Persistent Point Feature Histograms for 3D Point Clouds (2008, methodology): “Robust feature descriptors such as moment invariants [4], spherical harmonic invariants [5], and integral volume descriptors [6] have been proposed as point features and used for registering partial scans of a model [7,6].”
Deformable Medical Image Registration: A Survey (2013, background): “[316] investigated the use of shape features in addition to the positional information when estimating the correspondences.”
RGB-D Mapping: Using Depth Cameras for Dense 3D Modeling of Indoor Environments (2010, methodology): “This approach has been applied to point color [13], geometric descriptors [28], and point-wise reflectance values [19].”
Rapid object indexing using locality sensitive hashing and joint 3D-signature space estimation (2006, methodology): “Due to the small support of the features, errors of the LSH method and the distortions in the descriptors produced by articulations and other imperfections in the data, we employ a relatively large number of candidate matches, Q ¼ 50 100, for each descriptor in the scene.”
Towards 3D Point cloud based object maps for household environments (2008, methodology): “Additionally, to further improve the quality of correspondences, a lot of efforts have been made into the area of feature selection [10,35], as well as including extra information such as colors [17] or normals [2] that could improve the correspondence problem.”
The three-dimensional normal-distributions transform : an efficient representation for registration, surface analysis, and loop detection (2009, methodology): “[92] have used other kinds of invariant surface features for ICP registration.”
Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration (2016, methodology): “[36] proposed the additional use of invariant feature descriptor distance; Johnson and Kang [37] exploited color distances to boost the performance.”
Manipulator and object tracking for in-hand 3D object modeling (2011, methodology): “This approach has been applied to color (Johnson and Kang, 1997), geometric descriptors (Sharp et al., 2002), and image descriptors (Lemuz-López and Arias-Estrada, 2006).”