Using image segmentation for evaluating 3D statistical shape models built with groupwise correspondence optimization

Gollmer, Sebastian T.; Kirschner, Matthias; Buzug, Thorsten M.; Wesarg, Stefan

doi:10.1016/j.cviu.2014.04.014

Cited by 9 publications

(7 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Davies et al [9] show that MDL outperforms state-of-the-art registration methods for medical datasets. Gollmer et al [13] compare different objective functions. They show that while the determinant of the covariance matrix is easier to optimize, the results are comparable to results produced by MDL.…”

Section: Related Workmentioning

confidence: 99%

“…For models of different compactness that describe the same data, the model with higher compactness and hence lower variance is favorable. It has been shown that minimizing the variance of a PCA model performs similarly to information theoretic approaches that aim at minimizing the description length of the model [13].…”

Section: Groupwise Correspondence Optimizationmentioning

confidence: 99%

“…Regularization: To avoid undersampling in these regions, Davies et al [10] approximate the integral of the continuous covariance matrix by weighting the points by their surrounding surface area. Since this does not always prevent the undersampling [13], as done in Burghard et al [6], we use a regularization within the objective function. The regularization term for each shape is a bi-Laplacian of the form…”

Section: Multilinear Objective Functionmentioning

confidence: 99%

See 2 more Smart Citations

A Groupwise Multilinear Correspondence Optimization for 3D Faces

Bolkart¹,

Wuhrer²

2015

2015 IEEE International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

The official version of this article is available on the IEEE websiteInternational audienceMultilinear face models are widely used to model the space of human faces with expressions. For databases of 3D human faces of different identities performing multiple expressions, these statistical shape models decouple identity and expression variations. To compute a high-quality multilinear face model, the quality of the registration of the database of 3D face scans used for training is essential. Meanwhile, a multilinear face model can be used as an effective prior to register 3D face scans, which are typically noisy and incomplete. Inspired by the minimum description length approach, we propose the first method to jointly optimize a multilinear model and the registration of the 3D scans used for training. Given an initial registration, our approach fully automatically improves the registration by optimizing an objective function that measures the compactness of the multilinear model, resulting in a sparse model. We choose a continuous representation for each face shape that allows to use a quasi-Newton method in parameter space for optimization. We show that our approach is computationally significantly more efficient and leads to correspondences of higher quality than existing methods based on linear statistical models. This allows us to evaluate our approach on large standard 3D face databases and in the presence of noisy initializations

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Groupwise Correspondence Optimizationmentioning

confidence: 99%

Section: Multilinear Objective Functionmentioning

confidence: 99%

See 1 more Smart Citation

A Groupwise Multilinear Correspondence Optimization for 3D Faces

Bolkart¹,

Wuhrer²

2015

2015 IEEE International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…Recent advances in in vivo imaging of anatomy and the wide spectrum of shape modeling applications have led to the development and distribution of open-source SSM tools, further enabling their use in an off-the-shelf manner. Evaluation of SSM tools has been performed using non-clinical applications such as image segmentation [15] and shape/deformation synthesis approaches [13]. However, to the best of our knowledge, little work has been done on the evaluation and validation of such tools as related to clinical applications that rely on morphometric quantifications.…”

Section: Introductionmentioning

confidence: 99%

On the Evaluation and Validation of Off-the-Shelf Statistical Shape Modeling Tools: A Clinical Application

Goparaju

Csécs

Morris

et al. 2018

Shape in Medical Imaging

View full text Add to dashboard Cite

Statistical shape modeling (SSM) has proven useful in many areas of biology and medicine as a new generation of morphometric approaches for the quantitative analysis of anatomical shapes. Recently, the increased availability of high-resolution in vivo images of anatomy has led to the development and distribution of open-source computational tools to model anatomical shapes and their variability within populations with unprecedented detail and statistical power. Nonetheless, there is little work on the evaluation and validation of such tools as related to clinical applications that rely on morphometric quantifications for treatment planning. To address this lack of validation, we systematically assess the outcome of widely used off-the-shelf SSM tools, namely ShapeWorks, SPHARM-PDM, and Deformetrica, in the context of designing closure devices for left atrium appendage (LAA) in atrial fibrillation (AF) patients to prevent stroke, where an incomplete LAA closure may be worse than no closure. This study is motivated by the potential role of SSM in the geometric design of closure devices, which could be informed by population-level statistics, and patient-specific device selection, which is driven by anatomical measurements that could be automated by relating patient-level anatomy to population-level morphometrics. Hence, understanding the consequences of different SSM tools for the final analysis is critical for the careful choice of the tool to be deployed in real clinical scenarios. Results demonstrate that estimated measurements from ShapeWorks model are more consistent compared to models from Deformetrica and SPHARM-PDM. Furthermore, ShapeWorks and Deformetrica shape models capture clinically relevant population-level variability compared to SPHARM-PDM models.

show abstract

“…These advantages are quite important especially to complex threedimensional shapes (the overall dimensionality of a 3D shape as a whole is usually tremendously higher compared with its 2D counterparts). That is why this class of techniques is still quite prevalent [14][15][16][17][18][19][20][21][22][23], even though many more accurate nonlinear modeling approaches have been worked out. One thing to be mentioned is that a majority of linear modeling techniques adopt Principal Component Analysis (PCA) [24] as a fundamental tool to realize linear shape representation.…”

Section: Introductionmentioning

confidence: 99%

Vertex‐level three‐dimensional shape deformability measurement based on line segment advection

Hancock²,

Xiao

et al. 2018

IET Computer Vision

View full text Add to dashboard Cite

Abstract:The measurement to the intrinsic deformability of arbitrary small-scale subdivision of a shape is an interesting meanwhile valuable research topic. Such measurement can be directly utilized as a reliable criteria to partition object shape into small components and then assist in accurate shape modeling and description. Compared with single global shape modeling, through constructing subdivision-based complex shape description, the accuracy and flexibility of shape representation can be significantly improved. In this paper, we propose a line segment advection based vertex-level three-dimensional shape deformability measuring method. This measuring strategy can highlight the deformability characteristics of each shape part in any scale and size, at the same time, quantitatively describe them. The measurement is realized mainly based on the advection of line segments connecting neighboring shape mesh vertices. For 3D shapes, since the line segment of triangular mesh facet directly reflects the minimal neighborhood relationships and mesh microstructure, its advection can capture the finest details of shape deformability. Then, after transferring those information into neighboring vertices, a vertex-level shape deformability measurement can be acquired. Besides the measuring method, in order to demonstrate its value to shape partitioning and piecewise shape modeling, a straightforward shape partitioning method based on the proposed measurement is introduced as well, which is implemented through unsupervised clustering. Extensive experiments on three publicly available databases are conducted to verify the effectiveness of proposed methods.

show abstract

Using image segmentation for evaluating 3D statistical shape models built with groupwise correspondence optimization

Cited by 9 publications

References 64 publications

A Groupwise Multilinear Correspondence Optimization for 3D Faces

A Groupwise Multilinear Correspondence Optimization for 3D Faces

On the Evaluation and Validation of Off-the-Shelf Statistical Shape Modeling Tools: A Clinical Application

Vertex‐level three‐dimensional shape deformability measurement based on line segment advection

Contact Info

Product

Resources

About