A new algorithm for non-rigid point matching

By Haili Chui, Anand Rangarajan. 2000

537 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

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

Combining self-supervised learning and imitation for vision-based rope manipulation (2017, methodology): “[26] used non-rigid registration [27] for transferring human demonstrated trajectories for rope manipulation.”
Shape Matching and Object Recognition Using Shape Contexts (2002, methodology): “Most closely related to our approach is the work of Gold et al. [19] and Chui and Rangarajan [9], which is discussed in Section 3.4.”
The dual-bootstrap iterative closest point algorithm with application to retinal image registration (2003, methodology): “The nal comparison is to the robust point matching (RPM) algorithm [ 17 ].”
Robust Registration of 2D and 3D Point Sets (2003, background): “The separation of correspondence and update remains true of EM-based approaches [15, 25, 5].”
A robust algorithm for point set registration using mixture of Gaussians (2005, methodology): “Following the approach in [4], we choose the thin-plate spline (TPS) to represent the non-rigid deformation.”
Fast geometric learning with symbolic matrices (2020, unknown): “These two regimes correspond to typical applications in computer vision [24, 38] and statistics [43, 46, 48], respectively.”
Non-rigid point set registration: Coherent Point Drift (2006, methodology): “Several non-rigid registration methods are introduced [4, 5].”
Matching shapes (2001, methodology): “Most closely related to our approach is the work of Rangarajan and collaborators [12, 71, which is discussed in Section 3.2.”
A Bayesian network framework for relational shape matching (2003, methodology): “And, in a move that is strikingly remi-1 niscent of our previous TPS-RPM approach to non-rigid point matching [3], the template point-set is repeatedly warped using a thin-plate spline.”
Efficient shape matching using shape contexts (2005, methodology): “Another set of methods compute correspondences between edg e points, such as the work of Carlsson [19], which uses order structure, and the work of Johnson and Hebert [20] and Chui and Rangarajan [21].”
Interpolating between Optimal Transport and MMD using Sinkhorn Divergences (2018, methodology): “The idea of adding an entropic barrier KL( · |α⊗β) to the original linear OT program can be traced back to Schrödinger’s problem (Léonard, 2013) and has been used for instance in social sciences (Kosowsky and Yuille, 1994; Galichon and Salanié, 2010) and computer vision (Chui and Rangarajan, 2000).”
Point Set Registration: Coherent Point Drift (2009, methodology): “[7], and its later variants [5], [8], [9].”
Matching with shape contexts (2000, methodology): “Chui and Rangarajan [10] use TPS in their robust point matching algorithm.”
Composite Templates for Cloth Modeling and Sketching (2006, unknown): “Although the TPS-ICP algorithm is a locally optimal algorithm , the likelihood models with region information in the sub-templates are quite robust.”
Shape Context: A New Descriptor for Shape Matching and Object Recognition (2000, methodology): “In this work, we use the thin plate spline (TPS) model, which is commonly used for representing flexible coordinate transformations [17, 6].”
Human Ear Recognition in 3D (2007, unknown): “Thin plate spline (TPS) transformation is a powerful tool for modeling the shape deformation and is widely used in shape matching [24], [35], [36], [37].”
Recognition of shapes by editing their shock graphs (2004, methodology): “A drawback of such methods [4], [9] is their sensitivity to partial occlusion and articulation of parts, mainly due to the use of an extrinsic reference frame, see Fig.”