Kenong Wu

Live cell image segmentation

A major requirement of an automated, real-time, computer vision-based cell tracking system is an efficient method for segmenting cell images. The usual segmentation algorithms proposed in the literature exhibit weak performance on live unstained cell images, which can be characterized as being of low contrast, intensity-variant, and unevenly illuminated. The authors propose a two-stage segmentation strategy which involves: 1) extracting an approximate region containing the cell and part of the background near the cell, and 2) segmenting the cell from the background within this region. The approach effectively reduces the influence of peripheral background intensities and texture on the extraction of a cell region. The experimental results show that this approach for segmenting cell images is both fast and robust.<<ETX>>

3-D shape approximation using parametric geons

This paper presents a new approach to 3D shape representation - approximating the shapes of object parts by a set of prescribed volumetric models using single- and multi-view range data. We define a new set of volumetric part models, called parametric geons. These are seven qualitative shapes, each of which is formulated by a restricted globally-deformed superellipsoid. Model recovery is performed by fitting all parametric geons to a part and selecting the best model for the part based on the minimum fitting residual. A newly-defined objective function and a fast global optimisation technique are employed to obtain robust model fitting results. Parametric geons provide a global shape constraint that allows model recovery to explicitly verify the resultant part descriptions. Through systematic experiments, we examine the efficiency of the objective function, the discriminative ability of parametric geons, the effects of object shape imperfection to model recovery, and the importance of multiview data for shape approximation. The experimental results demonstrate that this approach can successfully recover qualitative shape models from object parts, especially when a part shape is not fully consistent with model shapes.

3D part segmentation using simulated electrical charge distributions

A novel approach to 3D part segmentation is presented. Beginning with range data of a 3D object, we simulate the charge density distribution over an objects surface which has been tessellated by a triangular mesh. We then locate the object part boundary at deep surface concavities by tracing local charge density minima. Finally, we decompose the object into parts at the part boundary points.

3D part segmentation: a new physics-based approach

We propose a novel approach to 3D part segmentation. From physics it is known that on the surface of a charged conductor, charge tends to accumulate at a sharp convexity and vanishes at a sharp concavity. Thus object part boundaries, which are usually denoted by a sharp surface concavity, can be detected by locating surface points exhibiting focal charge minima. Beginning with multiview range data of a 3D object, we simulate the electrical charge distribution over an objects surface which has been tessellated by a triangular mesh. We detect the deep surface concavities by tracing local charge density minima and then decompose the object into parts at these points.

2D shape segmentation: a new approach

We propose a novel approach to 2D shape segmentation by simulating the electrical charge density distribution on a 2D contour. The method is unique and invariant with respect to scale, translation and rotation. It computes significant 2D shape features by considering global data and is easily extensible to the 3D case.

Parametric geons: a discrete set of shapes with parameterized attributes

We propose parametric geons as a volumetric description of object components for qualitative object recognition. Parametric geons are seven qualitative shape types defined by parameterized equations which control the size and degree of tapering and bending. The models provide global shape constraints which make model recovery procedures robust against noise and minor variations in object shape. The surface characteristics of parametric geons are discussed. The properties of parametric geons and conventional geon models are compared. Experiments fitting parametric geons to multiview data using stochastic optimization were performed. Results show that unique descriptions of single-part objects with minor shape variations can be obtained with the parametric geon models.

Segmenting 3D Objects into Geons

We propose a new approach to 3D object segmentation and description. Beginning with multiview range images of a 3D object, we segment the object into parts at deep surface concavities. Motivated by physics, we detect these concavities by locating surface points where the simulated electrical charge density achieves a local minimum. The individual parts are then described by parametric geons. The latter are defined as seven distinctive volumetric shapes characterized by constrained superellipsoids, with deformation parameters controlling the tapering and bending. We obtain a unique part model by fitting all parametric geons to each part and classifying the fitting residuals. The advantage of our classification approach is its ability to approximate the shape of an object, even one not composed of perfect geon-like parts. The resulting 3D shape description is a prerequisite for generic object recognition.

Shape approximation: from multiview range images to parametric geons

We have studied the problem of deriving object part approximations by a new set of distinct volumetric shape types called parametric geons from multiview and single-view range data. This is accomplished by fitting the models to range data of single-part objects and then classifying the fitting residuals. We investigate how the number of object views can affect the ultimate shape approximation. Experimental results show that qualitative shape information can be recovered using data taken from single general views, and that multiview data significantly improve the accuracy of the quantitative model information.

SIGNAL-TO-SYMBOL MAPPING FOR LASER RANGEFINDERS

A new approach for computing qualitative part-based descriptions of 3D objects is presented. The object descriptions are obtained in two steps: Object segmenta-tion into parts and part model identiication. Beginning with single-or multi-view range data of a 3D object, we simulate the charge density distribution over an objects surface which has been tessellated by a triangular mesh. We detect the deep surface concavities by tracing local charge density minima and then decompose the object into parts at these points. The individual parts are then modelled by parametric geons. The latter are seven qualitative shapes, each of which is formulated by a restricted globally deformed superellipsoid. Model recovery is performed by tting all parametric geons to a part and selecting the best model for the part, based on the minimum tting residual. A newly deened objective function and a fast global optimisation technique are employed to obtain robust model tting results. Experiments demonstrate that this approach can successfully recover qualitative shape models from input data, especially when part shapes are not fully consistent with model shapes. The resultant object descriptions are well suited for symbolic reasoning and fast object recognition.