Soo-Mi Choi

Volumetric object reconstruction using the 3D-MRF model-based segmentation [magnetic resonance imaging]

A number of segmentation algorithms have been developed, but those algorithms are not effective on volume reconstruction because they are limited to operating only on two-dimensional (2-D) images. In this paper, we propose the volumetric object reconstruction method using the three-dimensional Markov random field (3D-MRF) model-based segmentation. The 3D-MRF model is known to be one of efficient ways to model spatial contextual information. The method is compared with the 2-D region growing scheme under three types of interpolation. The results show that the proposed method is better in the aspect of image quality than other methods.

An affective user interface based on facial expression recognition and eye-gaze tracking

This paper describes a pipeline by which facial expression and eye-gaze of the user are tracked, and then 3D facial animation is synthesized in the remote place based upon timing information of the facial and eye movement information. The system first detects a facial area within the given image and then classifies its facial expression into 7 emotional weightings. Such weighting information, transmitted to the PDA via a mobile network, is used for non-photorealistic facial expression animation. It turns out that facial expression animation using emotional curves is more effective in expressing the timing of an expression comparing to the linear interpolation method. The emotional avatar embedded on a mobile platform has some potential in conveying emotion between peoples via Internet.

Picture navigation using an ambient display and implicit interactions

There is increasing demand for ubiquitous displays that react to a users actions. We propose a method of navigating pictures on an ambient display using implicit interactions. The ambient display can identify the user and measure how far away they are using an RFID reader and ultrasonic sensors. When the user is a long way from the display, it acts as a digital picture and does not attract attention. When the user comes within an appropriate range for interaction, the display shows pictures that are related to the user and provides quasi-3D navigation using the TIP (tour into the picture) method. In addition, menus can be manipulated directly on a touch-screen or remotely using an air mouse. In an emergency, LEDs around the display flash to alert the user.

Composition of MPEG-7 visual descriptors for detecting adult images on the internet

We describe a similarity based adult image detection technique (SID) exploiting composed MPEG-7 visual descriptors. The technique with a large set of training adult image database and a smaller set of training non-adult image database is practically useful in detecting adult images with little false negatives. SID achieved 99% correct detections with 23% false positives when experimented with a database containing 1,300 training non-adult images, and 93.5% correct detections with 8.4% false positives when experimented with a database containing 12,000 training non-adult images. 9,900 training adult images are used for both experiments. Given a query, ten most similar images are retrieved. If majority of the retrieved are adult images, then the query is determined to be an adult image. Otherwise, it is determined to be a non-adult image. SID can detect adult Internet content with the aid of text filtering system as described in the later section.

Modelling, visualization, and interaction techniques for diagnosis and treatment planning in cardiology

Abstract Due to the development of new imaging devices, which produce a large number of tomographic slices, advanced techniques for the evaluation of large amounts of data are required. Therefore, computer-supported extraction of dynamic 3-D models of patient anatomy from temporal series is highly desirable. Since the diagnostician must be able to quickly make rational decisions based on the models, a high degree of accuracy is required within a minimum amount of time. We present modelling and visualization techniques that are realized within the Cardiac Station. Results for the application of these techniques to cardiac image data are given. In addition to providing information about the patients morphology, functional parameters can be derived from the data and visualized together with the model. In order to verify the model with the original image data and to plan for real intervention, interaction techniques are presented.

Subsurface scattering using splat-based diffusion in point-based rendering

Point-based graphics has gained much attention as an alternative to polygon-based approaches because of its simplicity and flexibility. However, current point-based techniques do not provide a sufficient rendering quality for translucent materials such as human skin. In this paper, we propose a point-based framework with subsurface scattering of light, which is important to create the soft and semi-translucent appearance of human skin. To accurately simulate subsurface scattering in multilayered materials, we present splat-based diffusion to apply a linear combination of several Gaussian basis functions to each splat in object space. Compared to existing point-based approaches, our method offers a significantly improved visual quality in rendering human faces and provides a similar visual quality to polygon-based rendering using the texture space diffusion technique. We demonstrate the effectiveness of our approach in rendering scanned faces realistically.

Motion visualization of human left ventricle with a time‐varying deformable model for cardiac diagnosis

We present a time-varying deformable model to visualize and analyze the motion of the left ventricle from a time series of 3-D images. The model is composed of a non-rigid body that deforms around a reference shape obtained from the previous time step. At each time step, the position and orientation of the left ventricle are extracted from the feature points of images. This information gives the position and orientation of the coordinate system attached to the non-rigid body. To compute a dense non-rigid motion field over the entire endocardial wall of the left ventricle, we introduce a 3-D blob finite element and Galerkin interpolants based on 3-D Gaussian, and use a physically based finite element method and a modal analysis. Then, cinematic attributes are visualized in pseudo colors on the reconstructed surface in order to help medical doctors in their interpretation of the data. Using the presented model, we estimate clinically useful quantitative parameters such as regional wall motion and ejection fraction. Experimental results are shown in a time series of X-ray angiographic images. Copyright ©2001 John Wiley & Sons, Ltd.

Morphometry of the hippocampus based on a deformable model and support vector machines

This paper presents an effective representation scheme for the statistical shape analysis of the hippocampal structure and its shape classification: Morphometry of the hippocampus. The deformable model based on FEM (Finite Element Method) and ICP (Iterative Closest Point) algorithm allows us to represent parametric surfaces and to normalize multi-resolution shapes. Such deformable surfaces and 3D skeletons extracted from the voxel representations are stored in the Octree data structure. And, it will be used for the hierarchical shape analysis. We have trained SVM (Support Vector Machine) for classifying between the control and patient groups. Results suggest that the presented representation scheme provides various level of shape representation and SVM can be a useful classifier in analyzing the statistical shape of the hippocampus.

Shape reconstruction from partially missing data in modal space

Abstract To reconstruct the shape of objects from incomplete point sets or noisy images, robust and accurate reconstruction methods are required. This paper presents a physics-based approach for automatically reconstructing three-dimensional shapes in a robust and proper manner from partially missing data. In modal space, an objects shape is described by modal deformations within each free vibration mode. In general, modes provide global-to-local ordering of shape deformation that allows us to select which types of deformations are to be used to reconstruct objects. By discarding high-frequency modes, the required computational time can be reduced without loss of accuracy. Moreover, the reconstructed shapes are relatively robust for partially missing data because of the stability of low-frequency modes. Unlike most existing approaches using a prototype with a priori fixed material properties, we calculate deformation modes directly from available data and allow the update of material properties to incorporate observations. As a result more improved shapes can be obtained. This new reconstruction algorithm is applied to point sets obtained from meshes and boundary contours in medical images.

Semantic interoperability of GIS and MCDA tools for environmental assessment and decision making

A promising approach to environmental assessment and decision making analyses is based on integrating Multicriteria Decision Analysis (MCDA) and GIS tools. Integration of GIS and MCDA tools can be potentially achieved through interoperability, where these tools can exchange relevant information to tackle a particular environmental problem. However, the problem of semantic heterogeneity caused by different meanings of data, terminologies, and models used in GIS and MCDA has been recognized as an obstacle in the interoperability of these tools. Conventionally, exchange of data between GIS and MCDA systems for environmental applications relies on prior knowledge to mediate meanings between the two components. This paper proposes an ontology-enabled framework for semantic interoperability of GIS and MCDA web services. In particular, this study has made significant contribution to environmental decision making by providing an interoperable framework to exchange environmental data with intended and unambiguous meanings between GIS and MCDA services. This paper proposes an ontology-enabled framework for semantic interoperability of GIS and MCDA components.It shifts the paradigm of environmental decision support systems from application-specific to a semantic interoperable framework.The approach uses a set of ontologies to address the challenges of semantic interoperability between GIS and MCDA tools.