Roman Durikovic

Dynamic contour: A texture approach and contour operations

The large morphometric variability in biomedical organs requires an accurate fitting method for a pregenerated contour model. We propose a physically based approach to fitting 2D shapes using texture feature vectors and contour operations that allow even automatic contour splitting. To support shrinkage of the contour and obtain a better fit for the concave parts an area force is introduced. When two parts of the active contour approach each other, it divides. The contour undergoing elastic deformation is considered as a set of masses linked by springs with their natural lengths set to zero. We also propose a method for automatic estimation of some model parameters based on a histogram of image forces along a contour.

Animation of Biological Organ Growth Based on L‐systems

In contrast with the growth of plants and trees, human organs can undergo significant changes in shape through a variety of global transformations during the growth period, such as bending or twisting. In our approach, the topology of a human organ is represented by a skeleton in the form of a tree or cycled graph. The length of skeleton growth can be simulated by an algebraic L‐system that also produces discrete events. The paper shows how to include global transformations into the formalism of L‐systems to obtain a continuous process. The shape of the organ is approximated by a number of ellipsoidal clusters centred at points on the skeleton. The proposed growth model of the organ continually responds to the positional changes of surrounding organs, thereby changing the organ shape locally. In our study, the stomach of a human embryo is used for the demonstration of organ development, and the methodology employed is also applicable to the animation of animal organs and their development.

GPU Rendering of the Thin Film on Paints with Full Spectrum

Spectrum-based rendering uses spectral distributions instead of just three RGB colors for representation of light sources and surface properties in rendering equation. Since, spectrum has a value at every visible wavelength, the spectrum-based rendering gives much accurate color computation compared to RGB-based rendering and it give us opportunity to simulate wavelength dependent phenomena and effects caused by spectrum difference. We introduce a novel framework for spectrum-based rendering on GPU (graphics processing unit) to compute local illumination. On this framework, the Phong reflectance model is implemented employing the spectral power distribution of light source and the spectral reflectance of surface simulating the color rendition of light sources and metamerism of surfaces. Multilayered thin film interference effects can also be handled within this framework with interactive speeds. Additionally, we propose the area light source defined by the spectral cube environment map and show the method for conversion of RGB environment map into a spectral one

Human hand model based on rigid body dynamics

In this paper, a method for dynamics simulation of articulated multi rigid body system is presented. We simulate a model of a hand, using constraints to express joints, motors and contacts between the bodies. The basic ideas of rigid body simulation and constraints are explained. Using contacts as a signal for motors, we create a physically valid motion of the hand model.

Preserving the Volume of Fluid Using Multi-phase Flow Approach

Methods to animate multiphase fluids are rarely researched, although such phenomena are common in our daily lives. The main focus of this research is to develop a suitable method to create realistic animation of multiphase flows, i.e. flow of fluid mixtures. Our method is based on Navier-Stokes equations, a set of physical equations that describe motion of fluid. A marker-and-cell (MAC) grid is used to solve the equations on discrete computational domain. To simulate several fluids, we propose the modification of volume-of-fluid (VOF) representation of fluid and integrate it into the multiphase-fluid approach. A mixture of fluids is treated as a single fluid having variable density and viscosity. This scheme allows two or more fluids having different densities and viscosities to be simulated simultaneously

Automated Usability Measurement of Arbitrary Desktop Application with Eyetracking

Nowadays in software development process more attention is paid to the final usability of the product. To achieve such usability we use various methods from user centered design up to the usability evaluation methods, requiring much attention from usability experts. The presense of these experts are needed both during capturing and analysing usability data, which eventually costs too much. We propose a tool for automated data capturing during user tests as well as a captured data analysis in order to evaluate the recorded interaction and guide the attention of software developers. Furthermore we provide a simple statistics of user tests as well as a means to browse recorded data with the interaction context extended with eye tracking data.

Shape-based calculation and visualisation of general cross-sections through biological data

The paper reports a solution to the problem of image interpolation for generating, with high precision, an arbitrary cross-section from a series of slices. Shape-based interpolation as proposed allows the interpolation process to be influenced by the shape of the object. It is accomplished first by creating reconstructed shapes of the objects from a series of slices. Correspondence vectors distributed along the object surface are then generated from the reconstructed shapes. These vectors ensure the correct image interpolation on the boundaries of objects. Next, the correspondence vectors are interpolated to every image point within the new cross-section in process, and the image points at the start and end of a vector are linearly interpolated. Places where the correspondence vectors cannot be (branching or holes in the object shape) are treated separately. Results are presented from biological data obtained using an optical microscope. The data indicate that the proposed method produces more accurate results than commonly used interpolation methods, although at the cost of increased computation.

Reconstructing a 3-D structure with multiple deformable solid primitives

This paper develops a sophisticated approach to a method for fitting of complex 3-D shapes to unsegmented sets of data. The approach features dynamic models consisting of parametrically defined solid primitives (such as superquadrics), global geometric transformations, global deformations and local finite element deformations. Our technique deals with a deformable model consisting of multiple primitives (parts) that can be used to represent even more complex shapes than are common in the natural world. The deformable model is extended to the case of unsegmented data, incorporating an extended version of a fuzzy clustering technique. The image or range data are transformed into forces acting on the model that deform and conform to the given data set while continually synthesizing nonrigid motion. An adaptive procedure updates a damping factor in accordance with the largest possible step in an implemented iteration process. We demonstrate the fitting process in experiments involving the extraction of the shape of the mouse embryos organs from microscopic sections.

Arbitrary cross-sections from biological data based on shape of organs

Techniques for generating a series of cross-sectional images, such as computed tomography (CT), magnetic resonance imaging (MRI) or direct sectioning, enable observation inside living organisms such as human or animal bodies, greatly contributing to medical and biological sciences. However, these techniques can interpret the cross-sections only in a prescribed direction and at certain large intervals limited by hardware. Furthermore, they do not allow immediate observation of arbitrary cross-sections. Straightforward linear interpolation can lead to artifacts in regions with large intensity gradients, particularly near the organ boundaries. This papers main purpose is to solve the problem of image interpolation for generating with high precision an arbitrary cross-section from a series of slices. This procedure is carried out by incorporating reconstructed organ shapes. The results are presented as biological data.

Optical-inertial Synchronization of MoCap Suit with Single Camera Setup for Reliable Position Tracking.

We propose a method for synchronization of an inertial motion capture suit and a single camera optical setup. Proposed synchronization is based on an iterative optimization of an energy potential in image space, minimizing the error between the camera image and a rendered virtual representation of the scene. For each frame, an input skeleton pose from the mocap suit is used to render a silhouette of a subject. Moreover, the local neighborhood around the last known position is searched by matching the silhouette to the distance transform representation of the camera image based on Chamfer matching. Using the combination of the camera tracking and the inertial motion capture suit, it is possible to retrieve the position of the joints that are hidden from the camera view. Moreover, it is possible to capture the position even if it cannot be captured by the suit sensors. Our system can be used for both real-time tracking and off-line post-processing of already captured mocap