Hwan-Gue Cho

Complex deformable objects in virtual reality

In this paper, we present a real-time animation technique for deformable objects based on mass-spring models in virtual reality environments. Many researchers have proposed various techniques for representing the motion and the appearance of deformable objects. However, the animation of deformable objects in virtual reality environments is still a hard problem. One of the most intensively studied deformable objects is virtual cloth. The difficulties in cloth animation mainly lie in the fact that cloth simulation easily tends to become unstable. Although the implicit method can make the simulation stable \citeBaraff98siggraph, it is still impossible to generate interactive animation when the number of mass-points is sufficiently large enough to represent realistic appearance. There have been a few efficient solutions for real-time animation \citeDesbrun99gi,Kang01tvc,Oshita01ca. However, the previous techniques for real-time cloth animation are not capable of generating the plausible motion and appearance because the methods are based on excessive approximations. This paper proposes an efficient technique that can generate realistic animation of complex cloth objects, and the technique makes it possible to integrate realistic deformable objects into the virtual reality systems without violating the interactivity of the system. The proposed method is based on the implicit integration in order to keep the system stable, and the solution of the linear system involved in the implicit integration is efficiently approximated in real-time.

GAME: A simple and efficient whole genome alignment method using maximal exact match filtering

In this paper, we present a simple and efficient whole genome alignment method using maximal exact match (MEM). The major problem with the use of MEM anchor is that the number of hits in non-homologous regions increases exponentially when shorter MEM anchors are used to detect more homologous regions. To deal with this problem, we have developed a fast and accurate anchor filtering scheme based on simple match extension with minimum percent identity and extension length criteria. Due to its simplicity and accuracy, all MEM anchors in a pair of genomes can be exhaustively tested and filtered. In addition, by incorporating the translation technique, the alignment quality and speed of our genome alignment algorithm have been further improved. As a result, our genome alignment algorithm, GAME (Genome Alignment by Match Extension), performs competitively over existing algorithms and can align large whole genomes, e.g., A. thaliana, without the requirement of typical large memory and parallel processors. This is shown using an experiment which compares the performance of BLAST, BLASTZ, PatternHunter, MUMmer and our algorithm in aligning all 45 pairs of 10 microbial genomes. The scalability of our algorithm is shown in another experiment where all pairs of five chromosomes in A. thaliana were compared.

Multiple genome alignment by clustering pairwise matches

We have developed a multiple genome alignment algorithm by using a sequence clustering algorithm to combine local pairwisegenome sequence matches produced by pairwise genome alignments, e.g, BLASTZ. Sequence clustering algorithms often generate clusters of sequences such that there exists a common shared region among all sequences in each cluster. To use a sequence clustering algorithm for genome alignment, it is necessary to handle numerous local alignments between a pair of genomes. We propose a multiple genome alignment method that converts the multiple genome alignment problem to the sequence clustering problem. This method does not need to make a guide tree to determine the order of multiple alignment, and it accurately detects multiple homologous regions. As a result, our multiple genome alignment algorithm performs competitively over existing algorithms. This is shown using an experiment which compares the performance of TBA, MultiPipMaker (MPM) and our algorithm in aligning 12 groups of 56 microbial genomes and by evaluating the number of common COGs detected.

Real-Time Animation of Complex Virtual Cloth with Physical Plausibility and Numerical Stability

Numerical instability has been a major obstacle to real-time cloth animation. Although the implicit method can make the simulation stable, it is still impossible to generate interactive animation when the geometric model of the virtual cloth is complex enough to represent realistic details. In this paper, we propose efficient animation techniques for the real-time animation of complex deformable objects. The proposed method exploits the stability of the implicit integration in order to use sufficiently large time steps for real-time environments, and obtains the approximate solution with optimized computation and storage. Unlike previous methods for real-time cloth animation, any severe simplification of the problem itself has been avoided in order to produce realistic motions of complex models. In addition, the proposed method deals with instability caused by a force derivative. This problem has been avoided in the previous work with excessive approximation or additional manipulation. However, the technique in this paper avoids instability with a minimal approximation, and does not increase any computational or geometric complexity. The method can be successfully integrated in various real-time environments, such as PC games and VR systems.

The Beta-Velocity Model for Simulating Handwritten Korean Scripts

The shape of oriental languages is different from that of western languages because the alphabet and the writing method are different. Thus western handwriting models cannot be applied directly to oriental languages although many handwriting models have been propose for alphabet-based languages. In this paper, we analyze Korean handwritten scripts and parameterize their characteristics. The parameters are used to generate various handwritten Korean scripts via our proposed model, the Beta-Velocity model. In order to generate Korean characters that consist of cursive strokes, we adopt a Beta distribution curve. The Beta distribution curve basis enables us to make the velocity model asymmetric. Combining these features, we can synthesize different Korean handwriting scripts by controlling the parameters. The proposed model can generate large amounts of data for on-line handwritten recognition systems, and it can support various graceful fonts for a commercial word-processor.

Visualization of whole genome alignment with LOD representation

The genome is the gene complement of an organism and it comprises the information of the entire genetic material of an organism. Many researchers use the whole genome alignment method to detect a genomic meaning between genomes. In this paper, we introduce a new method for whole genome alignment with LOD(Level-of-Detail) representation. It helps us to understand a relationship between two genomes and determine candidate sets from the whole genome alignment result.