A new digital media ear started a couple of months ago when 10 major cities in USA first boradcasted the all digital HDTV on the air. Digital video compression has been studied for decades but only recently, it becomes practical and is used in our daily life. We give an overview of the progress of video compression in recent years, particularly, the ones related to international standards. Through examples, we like to introduce this fantastic subject to the readers who were not familiar with it before.

The forthcoming century will be the "era of cyber society" . In a virtual environment, a cyber character plays the role of an agent for the human being. In this paper, we introduce active researches of industries, universities, and research institutes in Korea. After Adam Soft Inc. presented a cyber singer,“Adam”, a few cyber characters have succeeded commercially through Internet, TV, CF, movies and etc. Some universities had their own characters to make fundamental researches in the fields of modeling, motion control, and VR. Research institutes have been funded recently to a cyber character related researches. This paper briefly reviews the recent research works, and future trends in Korea. in Korea.

We present a practical method for removing biotches and restoring their mission data. To detect blotches, we employ a robust approach of local analysis of spatiotemporal anisotropic brightness continuity Our approach uses first-order spatiotemporal directional derivatives to select the smoothest direction for each examined pixel, and puts out the incorruption probability that he examined pixel may not be corrupted by blotches. As the restoration filter, were employ a spatiotemporal fuzzy filter whose response is adaptively controlled according to a fuzzy rule defined by the incorruption probability. The fuzzy filter is composed of the two different filter of the identity filter and the spatiotemporal directional-weighted-mean filter, and will put out an intermediate value between the original input brightness and the directional-weighted-mean brightness. We design the fuzzy rule in advance by a standard supervised learning fuzzy rule in advance by a standard supervised learning method. The computer simulations are presented.

To clarify the motion characteristics of free-style Okinawan dance“Kachaasi”, first the subjective impression was quantitatively evaluated with semantic differential technique to cluster its types. Then, the contingency of joint rotation in shoulder, elbow and wrist joints was examined with multivariate autoregressive model. The time-series data of positions and angels of three joints were calculated according to the deforming conditions and shielding directions of the ring lights. As the results, in an excellent dancer, the motions of shoulder and elbow were highly synchronized and smoothly controlled. The low-frequency output of the shoulder and elbow were mutually interacted. Meanwhile, the wrist behaved independently of other joints' rotation.

In this paper, the general formula of disparity estimation based on Bayesian Maximum A Posteriori (MAP) algorithm is derived. The generalized formula is implemented with the plane configuration model and applied to multiple baselined stereograms. The probabilistic plane configuration model consists of independence and similarity among the neighboring disparities in the configuration. The independence probabilistic model reduces the computation and guarantees the discontinuity at the object boundary region. The similarity model preserves the continuity or the high correlation of disparity distribution. In addition, we propose a hierarchical scheme of disparity compensation in the application to multiple-view stereo images. According to the experiments, the derived formula and the proposed estimation algorithm outperformed other ones. The proposed probabilistic models are reasonable and approximate the pure joint probability distribution very well with decreasing the computations to O(n(D)) from O(n(D)4) of the generalized formula. And, the hierarchical scheme of disparity compensation with multiple-view stereos improves the performance without any additional overhead to the decoder.

This paper proposes a new region-oriented coding scheme of still images. The scheme first segments an image into homogeneous regions by making use of Voronoi diagrams. Then luminance components in each Voronoi region are encoded using a transform coding technique which employs the Schmidt's orthogonalization method to produce orthonormal basis-functions for arbitrarily shaped regions. In this technique, coding efficiency depends on the properties of initial vectors to be orthogonalized. Therefore we adaptively select a set of initial vectors as well as the order in which these vectors are orthogonalized. Simulation results indicate that the proposed coding scheme achieves higher coding performance with much less computation than the KLT-based scheme which we reported formerly.

Information about camera operating such as zoom, focus, pan, tilt and tracking is useful not only for efficient video coding, but also for content-based video representation. A camera operating parameter measurement system designed specifically for these applications is therefore developed. This system, implemented in real time and synchronized with the video signal, measures the precise camera operating parameters. We calibrated the camera lens using a camera model that accounts for redial lens distortion. The system is then applied to infer image motion from pan and tilt operating parameters. The experimental results show that the inferred motion coincides with the actual motion very well, with an error of less than 0.5 pixel even for large motion up to 80 pixels.

In this paper, a novel gray-scale lossless image coder combining context-based minimum mean squared error (MMSE) prediction and entropy coding is proposed. To obtain context of prediction, this paper first defines directional difference according to sharpness of edge and gradients of localities of image data. Classification of 4 directional differences forms“geometry context”model which characterizes two-dimensional general image behaviors such as directional edge region, smooth region or texture. Based on this context model, adaptive DPCM prediction coefficients are calculated in MMSE sense and the prediction is performed. The MMSE method on context-by-context basis is more in accord with minimum entropy condition, which is one of the major objectives of the predictive coding. In entropy coding stage, context modeling method also gives useful performance. To reduce the statistical redundancy of the residual image, many contexts are preset to take full advantage of conditional probability in entropy coding and merged into small number of context in efficient way for complexity reduction. The proposed lossless coding scheme slightly outperforms the CALIC, which is the state-of-the-art, in compression ratio.

In this paper, we propose a method for reducing the blocking artifact in high-compressed JPEG images, which is based on minimization of block boundary discontinuity. We analyze the statistical characteristic of dequantized block boundary pixel difference values, classify all blocks in a JPEG-decoded image into one of four classes using that characteristics, and weight the compensating signal of each pixel estimated by the minimum block boundary discontinuity. Four weighting factors are selected based on the image content such as edge/nonedge, each of which is assigned to one of four classes. The proposed method can provide better visual look than any other method referred here by reducing the blocking artifact in the low as well as the high frequency regions.

The authors have considered the 1/f fluctuation of vial rhythm with $1/f\beta$ spectrum of $\alpha$ wave in relation to the invigoration for the learning memory by paid their attention to the hippocampus protocol in this paper. At the first clinical experiment, the data of the remembrance test at short period is able to make as the foundation of the repeat memory. It can replace this memory with long period memory through the hippocampus by the superposition of the same memory-nerve circuits.

Recently computer can make cyberspace to walk through by an interactive virtual reality technique. An a avatar in cyberspace can bring us a virtual face-to-face communication environment. In this paper, an avatar is realized which has a real face in cyberspace and a multiuser communication system is constructed by voice transmitted through network. Voice from microphone is transmitted and analyzed, then mouth shape and facial expression of avatar are synchronously estimated and synthesized on real time. And also an entertainment application of a real-time voice driven synthetic face is introduced and this is an example of interactive movie. Finally, face motion capture system using physics based face model is introduced.

This paper proposes an evolutionary model for automatically generating artificial creatures of various shapes and colors according to insect ecology. This model offers a novel way to naturally evolve the shapes and colors of artificial creatures. The evolutionary model used in our research is based on Genetic Algorithms (GA). In this paper, artificial Computer Graphics(CG) creatures develop into various shapes and colors according to the evolutionary model. Later, they can be used as CG animated characters. This model also solves the problem of reducing the time and labor cost for mass production of various characters. It could be used in such areas as the cavalry battle scene in Disney's animation, “Mulan”. Our approach has two steps. At first, artificial creatures move according to information gathered form the five senses. This information is also used for generating the shapes of the five sense organs[1]. Then, based on the GA, evolutionary mode[2], we prepare prototype creatures, which evolve into various shapes and different colors in alternating generations. Finally, our evolutionary model successfully generates various character shapes and colors automatically.

In this paper, we address practical methods for calculating camera lens distortion for real time applications. Although the lens distortion problem can be easily ignored for constant-parameter lenses, in the field of real-time camera calibrations, for zoom lenses a large number of calculations are needed to calculate the distortion. However, if the distortion can be calculated independently of the other camera parameter, we can easily calibrate a camera without the need for a large number of calculations. Based on Tsai's camera model, we propose two different methods for calculating lens distortion. These methods are so simple and require so few calculations that the lens distortion can be rapidly calculated even in real-time applications. The first method is to refer to the focal length - lens distortion Look Up Table(LUT), which is constructed in the initialization process. The second method is to use the relationship between the feature points found in the image. Experiments were carried out for both methods, results of which show that the proposed methods are favorably comparable in performance with non-real full optimization method.

In this paper, we propose a method of image-based three-dimensional modeling for realistic facial expression. In the proposed method, real human facial images are used to deform a generic three-dimensional mesh model and the deformed model is animated to generate facial expression animation. First, we take several pictures of the same person from several view angles. Then we project a three-dimensional face model onto the plane of each facial image and match the projected model with each image. The results are combined to generate a deformed three-dimensional model. We use the feature-based image metamorphosis to match the projected models with images. We then create a synthetic image from the two-dimensional images of a specific person's face. This synthetic image is texture-mapped to the cylindrical projection of the three-dimensional model. We also propose a muscle-based animation technique to generate realistic facial expression animations. This method facilitates the control of the animation. lastly, we show the animation results of the six represenative facial expressions.

MPEG-4 SNHC FBA(face and body animation) group is going to standardize the MPEG-4 system for low-bit rate communication with the implementation and animation of human body and face on virtual environment. In the first version of MPEG-4 standard, only the face object will be implemented and animated by using FDP (face definition parameter) and FAP(facial animation parameter), which are the abstract parameters of human face for low-bit rate coding. In this paper, MPEG-4 SNHC face object and it's animation were implemented based on the computer graphics tools such as VRML and OpenGL.

This paper describes TVML (TV program Making Language) for automatically generating television programs from text-based script. This language describes the contents of a television program using expression with a high level of abstraction like“title #1”and“zoom-in”. The software used to read a script written in TVML and to automatically generate the program video and audio is called the TVML Player. The paper begins by describing TVML language specifications and the TVML Player. It then describes the“external control mode”of the TVML Player that can be used for applying TVML to interactive applications. Finally, it describes the TVML Editor, a user interface that we developed which enables users having no specialized knowledge of computer languages to make TVML scripts. In addition to its role as a television-program production tool. TVML is expected to have a wide range of applications in the network and multimedia fields.

Homogeneous but distinct visual objects having low-contrast boundaries are usually merged in most of the segmentation algorithms. To alleviate this problem, an efficient image segmentation algorithm based on a bottom-up approach is proposed by using spatial domain information only. For initial image segmentation, we adopt an efficient marker extraction algorithm conforming to the human visual system. Then, two region-merging algorithms are successively applied so that homogeneous visual objects can be represented as simple as possible without destroying low-contrast real boundaries among them. The resultant segmentation describes homogeneous visual objects with few regions while preserving semantic object shapes well. Finally, a size-based region decision procedure may be applied to represent complex visual objects simpler, if their precise semantic contents are not necessary. Experimental results show that the proposed image segmentation algorithm represents homogeneous visual objects with a few regions and describes complex visual objects with a marginal number of regions with well-preserved semantic object shapes.

Template matching is one of the most often used techniques for machine vision applications to find a template of size M$\times$M or subimage in a scene image of size N$\times$N. Most template matching methods, however, require pixel operations between the template and the image under analysis resulting in high computational cost of O(M2N2). So in this thesis, we present a two stage template matching method. In the first stage, we use a novel low cost feature whose complexity is approaching O(N2) to select matching candidates. In the second stage, we use conventional template matching method to find out the exact matching point. We compare the result with other methods in terms of complexity, efficiency and performance. Proposed method was proved to have constant time complexity and to be quite invariant to noise.

With the recent evolutionary progress in network and terminal technologies, current videoconference systems are in the level of practical use. The existing systems, however, would not be fully useful for decision making process since eyes of users can not always be in contact that makes it difficult to read the expression of participants. In this paper, we discuss the principle and the impact of“MONJUnoCHIE System”a videoconference system with eye contact using a special display device called“Glass Vision”.

We have started the field test of the VOD service since April 1996, in Huis Ten Bosch which is one of the famous theme parks in Japan. In three years, total about 100 thousands people have used this VOD system and evaluated it. This test verifies that the VOD system which provides moving picture by MPEG2 via broad bandwidth ATM network has already reached the level of actual use. And also we get the know-how to construct contents, or how to manage network resources and server resources, and we have proposed them.

We describe a new videophone system for use both in intranet and internet environment. Video coding method for the videophone is based on the emerging MPEG-4 standard. One of the new functionalities is coding of video objects rather than video frames. Due to the object-based coding, the resulting videophone provides better quality and error robustness through intra/interne compared with the existing H.263-based videophones which employ frame-based video coding. The system is implemented in pure software on general-purpose multimedia personal computers and makes video telephony possible only with minimal hardware additions such as a set of video camera and video acquisition board. For the rest of the system, we used G.723.1 for speech, MPEG-4, synchronization layer (SL) for audio-visual synchronization, and RTP and RTCP over UDP for stream packet delivery. Call setup is performed using TCP. In a LAN environment, it currently provides superb audio-visual quality at rate as high as 15 frames/sec, which is still under continuing improvement as the system optimization progresses.

In computer graphics, the modeling and simulation of flames is a challenging problem. In this paper, we propose an approach for the simulation of a fire disaster. We use particle systems to describe the dynamic behavior fire. The illumination of dynamic flame is rendered by progressive radiosity algorithm.

Though a demand for 3-D CG (Three-Dimensional Computer Graphics) drastically increases, traditional modeling system for virtual objects are very difficult to handle. Applying VR (Virtual Reality) technology is, however, able to make the modeling systems more efficiently and intuitively. In this paper, traditional modeling operations are analyzed and their problems are clarified. We discuss what can be improved with VR technology. Then, a guideline of merging VR technology into the modeling system is described. A new modeling system is proposed using a CAVE system which is utilized in the latest VR technology. This proposed system aids in creating 3-D objects with high immersion in VEs (Virtual Environments)

In VEs (virtual environments) such as a CAVE system, there are three important operations : executive command decision, object selection and 3-D (three-dimensional) pointing. It is necessary to implement these operations in VEs intuitively and accessibly. In CAVE, it is possible for examines to walk and change their viewpoints freely. Then, the input devices which have excellent portability and rich expression are desired. Speech input satisfies both requirements. It is, however, very difficult for the speech input to indicate an exact point in 3-D space. Therefore, an extendable virtual stick is employed and it supports speech input. This paper proposes a user friendly interface using speech and a virtual stick in CAVE system. In this paper, several applications appropriate for the proposed interface are developed. Some problems are pointed out from the applications.

Recently a study of Virtual-Reality becomes active. Expectation to 3-D reproduction technology rises with it. Electro-holography using Spatial light modulator (SLM) attracts attention, because natural 3-D animated image is provided. In this paper recent development of 3D-TV by Electro-holography is reported. We consider the conditions to holographic TV such as input, transmission and display systems. Also we consider about fast hologram calculation because It is important to execute hologram calculation at high speed to realize the real time electro-holography. Then we consider the possibility of real time holography.

We have been making researches on 3-D displays using computer-generated holograms(CGHs). Our CGHs are binary Fresnel holograms that reconstruct point light sources and are recorded by using high resolution laser printers (image setters). We use an image setter with a resolution of 5080 dots per inch. It is possible to reconstruct CGHs with light-emitting points. As the resolution of the image setter is not so high, it is better to use a spherical wave as a reference beam. We considered the recordable points objects are restricted by the low resolution, and proposed the multiplex type hologram to reduce the number of point objects recorded in the unit area of the CGH. We proposed a method to make computer-generated color hologram which could reconstruct color point light sources, by combining RGB color filters with the stripe CGHs corresponding to each color. We considered two kinds of gradation method on our binary CGHs. In this paper, we propose a multiple reconstruction method for improving the narrow viewing field.

Radiosity method is a global illumination model for image synthesis. It computes all energy interactions among diffuse elements in a virtual environment. One of the major drawbacks if its time consuming computation. Existing radiosity algorithms for static scene is difficult to be applicable to dynamic environments. In this paper we proposed an hierarchical scene partition scheme to speedup the link update computations in the dynamic environments. Since the proposed spatial data structure is global, it not only can be used to speedup the culling of non-affected links after geometry change, but also can be used to accelerate the subsequent visibility computation. Several empirical tests are given to show the efficiency of our improved algorithm.

InterSpaceTM is an advanced networked virtual reality system that presents shared three-dimensional computer graphics (CG) worlds via the Internet where multiple users can enjoy synchronous communications with voice, video and text. Users can control their avatars as a surrogate interface. In InterSpace users can walk around and interact with other people and interact with contents. In this paper, we describe the function-distributed architecture used in InterSpace. The architecture enables flexible configuration of server functions and load distribution. It also allows users to select media and client PCs to switch servers dynamically.

When we see a binary black and white image, it appears. to our visual sense, to become clearer because of a pseudo halftone. This is because when we don't notice of all the details of the image, instead we see from a more global standpoint. We presented a theory and experimental results for layered model which extended external world, retina and brain of the two dimensional visual model. This paper propose the objective evaluation to coincide with the subjective evaluation of the human in ordered to evaluate the relative merits of the various modulation method. We obtain experimentally the restored images and the measure of approximation of the density four division and ordered dither method. The measure of approximation present s the objective evaluation scale to coincide with the subjective evaluation of the two modulation images.

Fractal coding is image compression techniques using one of image characteristics self-transformability. In fractal image coding, the encoding process is to select the domain block similar to a range block. The reconstructed image quality of fractal image coding depends on similitude between a range block and the selected domain block. Domain block similar to a range blocks. In fact, the error of the reconstructed image adds up the generated error in encoding process and the generated error in decoding process. But current domain block estimating function considered only the encoding error. We propose a domain block estimating function to consider not only the encoding error but also the decoding error. By computer simulation, it was verified to obtain the high quality reconstructed image.

MPEG-2 transcoding scheme is proposed as scalable transmission system for a video-on-demand system is described. In this system video contents encoded by MPEG-2 are transmitted to receiver terminals via communication lines that have different bit rate. A transcoding technique that directly converts a bit stream encoded by MPEG-2 into a lower coding rate stream without decoding is proposed. The transcoder called Trampeg includes size reduction and frame drop approach. The experimental results show that an MPEG-2 stream of 0 Mbps is able to be transcoded into a stream of 1.5 Mbps-300 Kbps.

This paper porposes and efficient coding scheme for remote medical communication systems, or“telemedicine systems”. These systems require a technique which is able to transfer large volume of data such as X-ray images effectively. We have already developed a hierarchical image coding and transmission scheme (HITS), which achieves an efficient transmission of medical images simply[1]. In this paper, a new coding scheme for HITS is proposed, which used hierarchical context modeling for the purpose of improving the coding efficiency. The hierarchical context modeling divides wavelet coefficients into several sets by the value of a correspondent coefficient in their higher class, or“a parent”, optimizes a Golomb-Rice (GR) code parameter in each set, and then encodes the coefficients with the parameter. Computer simulation shows that the proposed scheme is effective with simple implementation. This is due to fact that a wavelet coefficient has dependence on its parent. As a result, high speed data transmission is achieved even if the telemedicine system consists of simple personal computers.

Facial model reconstruction and surgical simulation are essential parts in the computer-aided surgical system. Plastic surgeons use it to design appropriate repair plans and procedures before actual surgery is operated. In this work, the exploration of 3-D metamorphosis to them presents new results in these two parts.

In Computer Graphics, Phong shading algorithm is essential and also sufficient for producing realistic images. In this paper, we propose an approach taking only two additions and one memory access for shading a pixel. A Phong-Shading table is used for storing the values of diffuse and specular components of the Phong reflection model. The intensity of a pixel can be obtained by table lookup. The performance of proposed method is almost the same as Gouraud shading.

Object segmentation fro natural video scenes has recently become one of very active research to pics due to the object-based video coding standard MPEG-4. Object detection and isolation is also useful for object-based indexing and search of video content, which is a goal of the emerging new standard, MPEG-7. In this paper, an automatic segmentation method of moving objects in image sequence is presented which is applicable to multimedia content authoring for MPEG-4, and two different segmentation approaches suitable for surveillance applications are addressed in raw data domain and compressed bitstream domains. We also propose an object-based video description scheme based on object segmentation for video indexing purposes.

A virtual studio is a new video production environment using interactive computer graphics and imaging media technology. The traditional chroma-keying with two-dimensional background is replaced by an advanced keying method with a dynamic computer-generated, three-dimensional background. We have developed a virtual studio system that is practical to use in the real production environment. It has not only essential features that are common among various commercial virtual studio systems, but also unique feature that help the producer to construct virtual studio sets and scenarios efficiently such as span graph, robust backup controller, and 3 dimensional character generator supporting all languages. Our virtual studio system was used in live broadcasting and proved that the system was practical enough. In this paper, we will introduce the structure and the major features of our system, called VdreamSet, and application examples to broadcasting.