Scale ratio ICP for 3D point clouds with different scales

Lin, Baowei; Tamaki, Toru; Raytchev, Bisser; Kaneda, Kazufumi; Ichii, Koji

doi:10.1109/icip.2013.6738457

Cited by 10 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For our object border idea, identical keypoints' position are significant because it can affect the perpendicular results between normal vectors. To this point, we use the same strategy as that of ISS (Intrinsic Shape Signatures) [19], but we combine point cloud keyscale [18] concept with detector to make it scale invariant.…”

Section: Keypoint Detectormentioning

confidence: 99%

“…, 2 . Here, the keyscale is to find the appropriate SPIN image width in terms of the minimum of cumulative contribution rate [18]. Next, we will use keyscale to confirm a certain search radius for ISS detector.…”

Section: Keyscalementioning

confidence: 99%

“…Keyscale [18] is a good way to register point clouds with different scale. This method estimates a keyscale to represent the scale size of target point cloud.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SIPF: Scale invariant point feature for 3D point clouds

Lin

Zhao

Tamaki

et al. 2015

2015 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a method for detecting ScaleInvariant Point Feature(SIPF) including 3D keypoints Detector and feature descriptor. To detect SIPF, we first estimate a keyscale for point cloud, and calculate the covariance matrix of each 3D point. Keypoints are the saliency points who have a fast change speed along with all principal directions. Then the descriptors are encoded based on the shape of a border or silhouette of an object to be detected or recognized. Experimental results with the Stanford datasets demonstrate that the proposed method can be effectively used for 3D point clouds expression.

show abstract

Section: Keypoint Detectormentioning

confidence: 99%

Section: Keyscalementioning

confidence: 99%

See 1 more Smart Citation

SIPF: Scale invariant point feature for 3D point clouds

Lin

Zhao

Tamaki

et al. 2015

2015 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…When dealing with the registration of 3D point sets with large scale difference, the scale factor can be recovered independently [17], or obtained as part of a global optimization [18] [19] problem. In [17] the authors propose to characterize the scale of a given point cloud by a set of cumulative contribution rate curves obtained by performing principal component analysis on spin images. A variant of ICP is then used to register the curves of two point clouds and recover the scale.…”

Section: Modelintegrationmentioning

confidence: 99%

Automatic video to point cloud registration in a structure-from-motion framework

Vidal

Piotto

Cordara

et al. 2015

2015 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

In Structure-from-Motion (SfM) applications, the capability of integrating new visual information into existing 3D models is an important need. In particular, video streams could bring significant advantages, since they provide dense and redundant information, even if normally only relative to a limited portion of the scene. In this work we propose a fast technique to reliably integrate local but dense information from videos into existing global but sparse 3D models. We show how to extract from the video data local 3D information that can be easily processed allowing incremental growing, refinement, and update of the existing 3D models. The proposed technique has been tested against two state-of-the-art SfM algorithms, showing significant improvements in terms of computational time and final point cloud density.

show abstract

“…Furthermore, the initial interactive (i.e., manual) alignment can influence the accuracy of point cloud registration [103]. Many modified versions of ICP algorithm have been introduced to address these limitations [104][105][106]. Despite improvements, however, the transformed source does not necessarily fit into the structure of the target data.…”

Section: Once Hyper-enhanced Regions On Lge-mri and Regions Of Low Bimentioning

confidence: 99%

Development of a Method for Registering Surface of Left Atrium determined by LGE-MRI to Electroanatomical Maps

Lee¹

View full text Add to dashboard Cite

Abnormal electrical conduction and excitability associated with fibrosis in the left atrium (LA) may serve as a substrate for atrial fibrillation (AF). Electroanatomical voltage mapping systems (EAMs) have become a dominant facilitator to treat AF with surgical ablation assisted by additional diagnostic imaging modalities. Importantly, AF has been associated with structural changes to the extracellular matrix of the myocardium, including increased collagen deposition -a process known as fibrosis. Late gadolinium-enhanced magnetic resonance imaging (LGE-MRI) may aid in guiding AF cardiac ablation therapy by determination of fibrosis in the LA. To locate fibrosis for cardiac ablation, however, precise registration between EAMs andLGE-MRI data is crucial. The purpose of this thesis was to develop a method for registering EAMs with late gadolinium-enhanced magnetic resonance image (LGE-MR) images. Twenty patients with persistent AF, who underwent magnetic resonance imaging scanning and EAMs prior to first-time catheter ablation, participated in the study. An experienced radiologist manually segmented the boundaries of the LA onLGE-MR images. In our registration pipeline, the first step involved manual alignment of the LGE-MR images to the left atrial surface on EAMs. The iterative closest point (ICP) algorithm was then applied to deform the left atrial surface determined using LGE-MRI to that of EAMs in an affine manner. We then applied a non-rigid iv ICP (NICP) algorithm to correct for non-linear deformations. The results demonstrate that NICP provided a substantial reduction in target registration error when compared to the use of affine ICP alone. v Dedication With warm and sincere thanks to Prof. Eranga Ukwatta for his superb guidance and feedback on technical image analysis; Dr. Rebecca E. Thornhill for her tireless research, insightful guidance on editing; Prof. Andy Adler for his supervision; Dr. Pablo Nery for initiating the research project; Dr. Robert DeKemp for offering his laboratory to support my research.Thanks also to the biomedical engineering laboratory colleagues who helped me go through the M.A.Sc graduate study with great understanding as well as for their friendship; Carleton University for giving me the opportunity to achieve M.A.Sc degree.

show abstract

Scale ratio ICP for 3D point clouds with different scales

Cited by 10 publications

References 17 publications

SIPF: Scale invariant point feature for 3D point clouds

SIPF: Scale invariant point feature for 3D point clouds

Automatic video to point cloud registration in a structure-from-motion framework

Development of a Method for Registering Surface of Left Atrium determined by LGE-MRI to Electroanatomical Maps

Contact Info

Product

Resources

About