Sparse Shape Representation Research Articles

Implicit Sparse Shape Representation: A Unified Framework for Object Segmentation and Recognition

Given a classified probabilistic shape dictionary, and an image with a shape similar to some of the elements in the dictionary, this letter introduces a sparse representation based framework with a twofold goal. First, to select a sparse shape combination from the dictionary that best represents the shape, and second, to accurately segment the image taking into account both the sparse shape combination and the image information. A new energy function that combines the region-based segmentation with sparse representation is introduced to accomplish both goals simultaneously. The experimental results show the superior segmentation and recognition capabilities of the proposed model.

Similarity estimation based on sparse spectral correspondence

We present a sparse spectral correspondence method for deformable shape analysis. Our method exploits randomized sampling for sparse shape representation. By choosing a random subset of points that preserve key properties of the entire data set, it allows one to run algorithms efficiently on a small sample. First we implement random row sampling of an undirected weighted graph matrix by “Lewis weights”, which can be viewed as statistical leverage scores of a reweighted matrix and used directly as sampling probabilities. Second, a sparse graph is constructed on selected sample points by using minimum spanning tree (MST). We then discover the meaningful structural correspondence based on TPS (thin-plate spline) approach in the spectral embedded space. Finally we show how we use the sparse spectral correspondence to implement similarity estimation for shape matching and classification for different topological shapes. A series of experimental results demonstrate that our method is accurate and robust for shape analysis.

Myocardium segmentation from DE MRI with guided random walks and sparse shape representation.

For patients with myocardial infarction (MI), delayed enhancement (DE) cardiovascular magnetic resonance imaging (MRI) is a sensitive and well-validated technique for the detection and visualization of MI. The myocardium viability assessment with DE MRI is important in diagnosis and treatment management, where myocardium segmentation is a prerequisite. However, few academic works have focused on automated myocardium segmentation from DE images. In this study, we aim to develop an automatic myocardium segmentation algorithm that targets DE images. We propose a segmentation framework based on both prior shape knowledge and image intensity. Instead of the strong request of the pre-segmentation of cine MRI in the same session, we use the sparse representation method to model the myocardium shape. Data from the Cardiac MR Left Ventricle Segmentation Challenge (2009) are used to build the shape template repository. The method of guided random walks is used to integrate the shape model and intensity information. An iterative approach is used to gradually improve the results. The proposed method was tested on the DE MRI data from 30 MI patients. The proposed method achieved Dice similarity coefficients (DSC) of 74.60 ± 7.79% with 201 shape templates and 73.56 ± 6.32% with 56 shape templates, which were close to the inter-observer difference (73.94 ± 5.12%). To test the generalization of the proposed method to routine clinical images, the DE images of 10 successive new patients were collected, which were unseen during the method development and parameter tuning, and a DSC of 76.02 ± 7.43% was achieved. The authors propose a novel approach for the segmentation of myocardium from DE MRI by using the sparse representation-based shape model and guided random walks. The sparse representation method effectively models the prior shape with a small number of shape templates, and the proposed method has the potential to achieve clinically relevant results.

A Preferred Curvature-Based Continuum Mechanics Framework for Modeling Embryogenesis

Mechanics plays a key role in the development of higher organisms. However, understanding this relationship is complicated by the difficulty of modeling the link between local forces generated at the subcellular level and deformations observed at the tissue and whole-embryo levels. Here we propose an approach first developed for lipid bilayers and cell membranes, in which force-generation by cytoskeletal elements enters a continuum mechanics formulation for the full system in the form of local changes in preferred curvature. This allows us to express and solve the system using only tissue strains. Locations of preferred curvature are simply related to products of gene expression. A solution, in that context, means relaxing the system’s mechanical energy to yield global morphogenetic predictions that accommodate a tendency toward the local preferred curvature, without a need to explicitly model force-generation mechanisms at the molecular level. Our computational framework, which we call SPHARM-MECH, extends a 3D spherical harmonics parameterization known as SPHARM to combine this level of abstraction with a sparse shape representation. The integration of these two principles allows computer simulations to be performed in three dimensions on highly complex shapes, gene expression patterns, and mechanical constraints. We demonstrate our approach by modeling mesoderm invagination in the fruit-fly embryo, where local forces generated by the acto-myosin meshwork in the region of the future mesoderm lead to formation of a ventral tissue fold. The process is accompanied by substantial changes in cell shape and long-range cell movements. Applying SPHARM-MECH to whole-embryo live imaging data acquired with light-sheet microscopy reveals significant correlation between calculated and observed tissue movements. Our analysis predicts the observed cell shape anisotropy on the ventral side of the embryo and suggests an active mechanical role of mesoderm invagination in supporting the onset of germ-band extension.

基于局部稀疏形状表示的医学图像分割模型

基于局部稀疏形状表示的医学图像分割模型

Object segmentation framework based on dictionary‐group and sparse shape representation

Given an input object whose shape is partly similar to some of the samples in the training set, a dictionary-group-based sparse model is introduced that can use the local information of those less similar shape neighbours to represent the object and guide the segmentation. The model follows from a new sparse energy function that combines a series of sparse local constraints with the fuzzy log-polar decomposition-based shape elements. Finally, a unified framework is built to connect the high-level shape representation with the low-level image segmentation. The model on the public datasets is tested, and the experimental results show the superior shape segmentation capabilities of the proposed model.

Implicit kernel sparse shape representation: a sparse-neighbors-based objection segmentation framework.

This paper introduces a new implicit-kernel-sparse-shape-representation-based object segmentation framework. Given an input object whose shape is similar to some of the elements in the training set, the proposed model can automatically find a cluster of implicit kernel sparse neighbors to approximately represent the input shape and guide the segmentation. A distance-constrained probabilistic definition together with a dualization energy term is developed to connect high-level shape representation and low-level image information. We theoretically prove that our model not only derives from two projected convex sets but is also equivalent to a sparse-reconstruction-error-based representation in the Hilbert space. Finally, a "wake-sleep"-based segmentation framework is applied to drive the evolutionary curve to recover the original shape of the object. We test our model on two public datasets. Numerical experiments on both synthetic images and real applications show the superior capabilities of the proposed framework.

Implicit kernel sparse shape representation: a sparse-neighbors-based objection segmentation framework

This paper introduces a new implicit-kernel-sparse-shape-representation-based object segmentation framework. Given an input object whose shape is similar to some of the elements in the training set, the proposed model can automatically find a cluster of implicit kernel sparse neighbors to approximately represent the input shape and guide the segmentation. A distance-constrained probabilistic definition together with a dualization energy term is developed to connect high-level shape representation and low-level image information. We theoretically prove that our model not only derives from two projected convex sets but is also equivalent to a sparse-reconstruction-error-based representation in the Hilbert space. Finally, a “wake–sleep”-based segmentation framework is applied to drive the evolutionary curve to recover the original shape of the object. We test our model on two public datasets. Numerical experiments on both synthetic images and real applications show the superior capabilities of the proposed framework.

Multi-Pose and Occluded Facial Landmark Localization Via Sparse Shape Representation

Automatic facial landmark localization is a challenging problem for real world images because of face pose variations and occlusions. This paper proposes a unified framework to robustly locate facial landmarks under different poses and occlusions. Instead of explicitly modeling the statistical point distribution, we use a sparse linear combination to approximate the observed shape, and hence alleviate the multi-pose problem. In addition, we use the sparsity constraint to handle outliers caused by occlusions. We also model the initial misalignment and use convex optimization techniques to solve them simultaneously and efficiently. We evaluated the proposed method extensively on both synthetic and real data. The experimental results are promising on handling pose variations and occlusions.

SU‐E‐J‐130: Automating Liver Segmentation Via Combined Global and Local Optimization

Purpose:The aim of this work is to develop a robust algorithm for accurate segmentation of liver with special attention paid to the problems with fuzzy edges and tumor.Methods:200 CT images were collected from radiotherapy treatment planning system. 150 datasets are selected as the panel data for shape dictionary and parameters estimation. The remaining 50 datasets were used as test images. In our study liver segmentation was formulated as optimization process of implicit function. The liver region was optimized via local and global optimization during iterations. Our method consists five steps: 1)The livers from the panel data were segmented manually by physicians, and then We estimated the parameters of GMM (Gaussian mixture model) and MRF (Markov random field). Shape dictionary was built by utilizing the 3D liver shapes. 2)The outlines of chest and abdomen were located according to rib structure in the input images, and the liver region was initialized based on GMM. 3)The liver shape for each 2D slice was adjusted using MRF within the neighborhood of liver edge for local optimization. 4)The 3D liver shape was corrected by employing SSR (sparse shape representation) based on liver shape dictionary for global optimization. Furthermore, H‐PSO(Hybrid Particle Swarm Optimization) was employed to solve the SSR equation. 5)The corrected 3D liver was divided into 2D slices as input data of the third step. The iteration was repeated within the local optimization and global optimization until it satisfied the suspension conditions (maximum iterations and changing rate).Results:The experiments indicated that our method performed well even for the CT images with fuzzy edge and tumors. Comparing with physician delineated results, the segmentation accuracy with the 50 test datasets (VOE, volume overlap percentage) was on average 91%–95%.Conclusion:The proposed automatic segmentation method provides a sensible technique for segmentation of CT images.This work is supported by NIH/NIBIB (1R01‐EB016777), National Natural Science Foundation of China (No.61471226 and No.61201441), Research funding from Shandong Province (No.BS2012DX038 and No.J12LN23), and Research funding from Jinan City (No.201401221 and No.20120109).