The particle method is widely used for CG simulation of water. However, it has not been widely used for the CG simulation of gases. One of the reasons for this is the difficulty in handling the boundary conditions in the simulation of gases. In this paper, I investigate the boundary condition for explosion, and suggest an "Upwind Inflow Boundary Condition" to determine the inflow position back to the upwind direction by considering the velocity of particle present near the open surfaces of the simulation space. This technique can automatically determine the particle position and velocity for the inflow while suppressing the inhibition of the particles' outflow.

View full abstract

Particle effects are vital components of computer graphics in modern computer games. While game developers have a choice of several different methods for particle effects on PCs and home consoles, there exist only few solutions for games in the fast growing smartphones/tablets market. This is not only because of the more than a magnitude lesser computational performance of the systems-on-a-chip used there, but especially due to their even much slower memory access, which renders nearly all approaches used on standard PCs unsuitable for smartphones/tablets. To overcome the bottleneck of the memory access, I suggest using a procedural approach, which will be described fully in this paper, with the example of real-time water. It is based on particle movement in 2D, but by applying physical forces directly to the particles rather than using a pressure-field like in e.g. the popular Navier-Stokes based methods. This has the advantage of avoiding the need for two different data structures, one for the pressure-field and one for the particles themselves, and thus reduces memory usage significantly. In this paper I will also present a simple way to introduce interactions with the particle effect as well as a comparison with a low-complex Navier-Stokes based approach. To the best of my knowledge, this is the first scientific work investigating particle effects for systems-on-a-chip, like e.g. smartphones/tablets.

View full abstract

For point cloud data obtained from 3D scanning devices, excessively large storage and long post-processing time are required. Due to this, it is very important to simplify the point cloud to reduce calculation cost. In this paper, we propose a new point cloud simplification method that can maintain the characteristics of surface shape for unstructured point clouds. In our method, a segmentation range based on mean curvature of point cloud can be controlled. The simplification process is completed by maintaining the position of the representative point and removing the represented points using the range. Our method can simplify results with highly simplified rate with preserving the form feature. Applying the proposed method to 3D stone tool models, the method is evaluated precisely and effectively.

View full abstract

In 3D CAD systems, it is not always possible to convert trimmed surfaces to the intended shapes upon data exchange between different CAD systems because the definition of trimmed surfaces is not unified among the CAD systems. If a shape can not import to a CAD system, the shape should be modified to suit to the system. But many problems to improve the shape arise. To solve the problems, it is effective to generate a new surface from the boundary curves and delete the invalid one. If insufficient information on the features of the original trimmed surface is applied, the intended surface cannot be estimated from the boundary curves. Even if surface can be estimated, the quality of the surface is low. We, therefore, focus on the tangent vectors of the original source surface in the U and V directions on the boundary curves of the trimed surface. In this paper, we present a new method to generate the new source surface by using the boundary curves and the features of the original source surface. In our method, at first, the first derivative vectors are calculated at the start and end points of each boundary curve and the division points on the curve segment of the source surface in the U and V directions. Next, the projection plane is defined whose origin is the maximum and minimum points of a boundary curve in the UV space and normal vector at the origin is the tangent vector of the original source surface. The offset direction is calculated by linear interpolation of the first derivative vectors in the U and V directions for each boundary curve, and the boundary curves are offset to the projection plane. Finally, a region enclosed by the offset four free-form curves is interpolated by a B-spline surface to create the new source surface.

View full abstract

In this paper, we propose and develop an interactive video system "GAYAIT UP". This system is an extension of our former system "GAYAIT". GAYAIT displays a lot of videos simultaneously, and a user can select one of them interactively and emphasize it. The former system a user's uses gestures or mouse operations to select a video. On the other hand, a white card is used to select a video, and the selected video is projected on the card in GAYAIT UP. As the result, the user can feel as if picking up a video from many ones and bringing it nearby. We describe a method for developing the system and we also propose applications of the system.

View full abstract

This paper proposes IFWB (Instant Fourth Wall Breaker) as a novel image-based modeling system that can provide a user with a feeling of presence in the world of image he/she is seeing. The user is required to input only a single image with a sense of depth (Fig. 1(a)), and then IFWB augments its depth without any further input and allows him/her to walk through the world of image freely (Fig. 1(b)). We mainly target at light and compact devices such as smartphones and tablets in order for the user to use the system readily just like a camera anytime and anywhere (Fig. 1(c)).

View full abstract

We propose a method for generating edge-hatching op-art that is represented by parallel or cross lines near edges from photograph. The proposed method is performed by using inverse laplacian filter. The effectiveness of our method is investigated experimentally. Variations in the edge-hatching op-art generated from the Lena image are obtained by varying the proposed method parameters, and edge-hatching op-arts are also generated from several other photographic images to assess their visual effects.

View full abstract

We propose a non-photorealistic rendering method for generating checkered pattern images which are checkered patterns in gray scale photographs. Our method generates checkered pattern images by an iterative process using inverse iris filter. The effectiveness of our method is investigated experimentally. Variations in the checkered pattern image generated from the Lena image are obtained by varying the inverse iris filter parameters, and checkered pattern images are also generated from several other photographic images to assess their visual effects.

View full abstract