Published papers

Understanding a volume dataset through a 2D display is a complex task because it usually contains multi-layered inner structures that inevitably cause undesirable overlaps when projected onto the display. This requires us to identify feature subvolumes embedded in the given volume and then visualize them on the display so that we can clarify their relative positions. This article therefore introduces a new feature-driven approach to previewing volumes that respects both the 3D nested structures of the feature subvolumes and their 2D arrangement in the projection by minimizing their occlusions. The associated process begins with tracking the topological transitions of isosurfaces with respect to the scalar field, in order to decompose the given volume dataset into feature components called interval volumes while extracting their nested structures. The volume dataset is then projected from the optimal viewpoint that archives the best balanced visibility of the decomposed components. The position of the optimal viewpoint is updated each time when we peel off an outer component with our interface by calculating the sum of the viewpoint optimality values for the remaining components. Several previewing examples are demonstrated to illustrate that the present approach can offer an effective means of traversing volumetric inner structures both in an interactive and automatic fashion with the interface.

In this paper, we propose a novel, particle-based flame simulation method, which relies on SPH to simulate the dynamics of flames by introducing a buoyancy term as an external force to the Navier-Stokes equations. The buoyancy term is directly affected by heat generated from a simple chemical reaction model. For efficiency, we adaptively re-sample particles around a flame. We demonstrate that it is possible to execute three-dimensional flame simulations in real-time for typical cases including transition from initial to stable states, and one-way coupling with solid. Our method offers reasonable results on CPU, while more realistic results can be obtained using GPGPU.

We have developed a hybrid wind tunnel, where 2D measurement-integrated (MI) simulation, which utilizes actual data acquired from real air flow behind a square cylinder, plays an important role in improving the accuracy of its numerical analysis. The wind tunnel requires an accompanying visual analysis tool with which we can effectively peer into the relationships between the actual and simulated flow fields. In this paper, we attempt to exploit the augmented reality-style display technique to that end. The basic idea is to instantaneously superimpose the computationally-visualized MI simulated pressure field onto the actual flow velocity structure physically-visualized with oil mist-traced streak lines. Space registration of these two visual sources is rather straight-forward since the fixed cylinder of the wind tunnel is easily identified, whereas visualizing the MI simulated pressure field can be characterized with its sophisticated scheme based on differential topology. Considering the facts that vortex centers are located at local minima in the pressure field, and each minimum is surrounded by a derived topological feature called ravine cycle, we can colorize the field adaptively and keep track of Karman vortex streets robustly, regardless of drastic change in the Reynolds number of the flow field.

In this paper, we propose a novel particle-based flame simulation method, which relies on SPH to simulate the dynamics of flame by introducing a buoyancy term as an external force to Navier-Stokes equation. The buoyancy term is directly affected by heat generated from a simple chemical reaction model. Good interchangeability allows us to easily couple our method with existing SPH source codes. For efficient execution, we adaptively re-sample particles around flame. We demonstrate three-dimensional simulations in real-time for typical cases including transition from initial to stable states, and one-way coupling with solid.

When visualizing large-scale particle systems, it is difficult to maintain adequate framerates because we have to render dynamic scenes with a large number of small spheres. For the control of trade-offs between the overall image quality and total rendering speed, we propose a new rendering scheme which uses a fast method based on shaded texture mapping and a high-quality implicit surface method in a combined way. The shaded texture mapping, which generates a pseudo-texture through alpha-blending a proper portion of template texture for shade and highlight onto a base spherical texture, can render a particle faster than the implicit surface method. However, a weakness of the texture mapping lies in its poor shading quality. In contrast, the implicit surface method is accurate enough for analyzing particle systems visually. Actual method to render each particle is decided according to the viewing distance; the high-quality method is chosen only when the distance is smaller than a threshold, to allow the user to observe the region of interest closely. We use a molecular dynamics simulation dataset to evaluate the effectiveness of our scheme empirically. In addition to this, we also consider the extensibility of our scheme in terms of framerate stability, scalability, and expressiveness.

In this paper, we propose a particle-based method for real-time flame rendering. In the field of visualization, particle-based volume rendering has been introduced for the purpose of previewing. Our method is a reasonably-simplified version of the particle-based volume rendering. Compared with previous flame rendering methods, both detail representation and controllability are sufficiently improved. In addition to this, it is easily implemented using standard APIs, and its fast enough rendering speed allows us to adjust environmental settings on the fly. Moreover, our method requires a few standard parameters as particle attributes, and thus it would be used in conjunction with existing particle simulators.

Seasonally-transient contrast between snow and mountain surfaces makes a big appeal to every viewer. Generally, when deciding how much snow covers the mountain, we need to take many climatological factors into account, and it is likely to complicate the rendering process. In this paper, we present a phenomenological model which easily depicts snow-covered mountains. Our method relies on our own surface descriptor, which is based primarily on Tangent Sphere Accessibility of smoothed height field, and also considers geometric features like azimuth direction and altitudes to a limited extent. The feasibility of the proposed method is illustrated through application to a real height map dataset.

Interpolating diffusion tensor fields is a key technique to visualize the continuous behaviors of biological tissues such as nerves and muscle fibers. However, this has been still a challenging task due to the difficulty to handle possible degeneracy, which means the rotational inconsistency caused by degenerate points. This paper presents an approach to interpolating 3D diffusion tensors in 2D planar domains by aggressively locating the possible degeneracy while fully respecting the underlying transition of tensor anisotropy. The primary idea behind this approach is to identify the degeneracy using minimum spanning tree-based clustering algorithm, and resolve the degeneracy by optimizing the associated rotational transformations. Degenerate lines are generated in this process to retain the smooth transitions of anisotropic features. Comparisons with existing interpolation schemes will be also provided to demonstrate the technical advantages of the proposed approach.

本論文では，山岳形状における積雪の様子を再現する現象学的モデルを提案する．ここでは，積雪の時間変化や山肌とのコントラストなどの美しさを手軽に表現し，しかも制御可能にすることを目的としている．山肌における積雪の度合いを決めるためには，その場所での気候や気象，雪の密度・粘性・固さなど，一般的に現実世界の多くの物理的ファクタを考慮する必要があるが，本手法では独自のサーフェス記述子を介することで積雪の現象学的再現を試みる．そのサーフェス記述子は，平滑化された高さ場におけるTangent Sphere Accessibility をベースに，方角や高度などの幾何学的条件を限定的に考慮している．本論文ではさらに，実際の標高データを用いた山岳形状に適用し，提案手法の視覚的効果を検証する．

Introducing distortions into perspective views is a popular technique to direct our visual attention to specific objects, as seen in hand-drawn illustrations and cartoon animations. This type of image expression, called nonperspective projection, is feasible in visual communication, because the human visual system can reconstruct the target three-dimensional (3D) scene correctly provided that the corresponding image distortions are within a certain perceptual tolerance. In this paper, we develop a perceptual approach to guiding the design of such nonperspective images by referring to the 3D perception induced by pictorial depth cues. We formulate an acceptable tolerance by investigating how we perceive image distortion according to the change in the configuration of depth cues. The obtained formulation is then incorporated into our new algorithm, with which we can automatically control plausible image deformation by simply modifying the positions and sizes of specific objects in a scene.

Splashing occurs when a liquid drop hits the solid or fluid surface at a high velocity. The drop after the impact spreads and forms a corona with a thickened rim, which first develops annular undulations and then breaks into secondary droplets. We have many chances to see splashes in our daily life, e.g., milk crown, splashing paint, and raindrops falling onto a pool, whose characteristics of deformation have a significant impact on the visual reality of the phenomena. Many experimental studies have been conducted to find criteria on when splashing would occur, but the physical mechanisms of splashing are still not completely understood. It was only recently discovered that ambient gas pressure is a principal factor for creating such a splash. In this paper, therefore, we newly incorporate the ambient gas pressure effect into the Navier-Stokes equations through SPH fluid simulation for representing more accurate splashing dynamics. Our experiments demonstrated that the new approach requires very little additional computing cost to capture realistic liquid behaviors like fingering, which have not previously been attained by SPH nor most schemes for fluid simulation.

変奏曲とは，主題とよばれる素材を示して，その後主題を変化させた変奏を複数並べていくクラシック音楽の形式の一つである．変奏は主題の各要素に対して行われるが，変奏様式によって主題がどのように変化しているかを判別するには音楽理論の知識が不可欠である．そこで本研究では，音楽知識をもたないユーザにも，変奏様式による主題の変化をわかりやすく提示することを目的として，変奏曲の構造を可視化する手法を提案する．入力にはMIDI(Musical　Instrument Digital Interface) データを用いて，変奏曲を3次元空間内に可視化し，構造をより理解しやすくするための対話的機能を提供するシステムを開発した．そこでは，各変奏様式と3 次元空間内のオブジェクトを対応付ける独自のマッピングを用いている．さらに独自のマッピングに関するユーザテストを行い，提案したマッピングの有効性が示された．

本論文では, 奥行きのある構図をもつ画像を1枚入力すると，ユーザ入力を必要とすることなく即座に3次元情報を付与し, ユーザ頭部の位置変化に合わせて変化する没入型映像を生成するシステムIFWBを提案する．実際のターゲット環境としてスマートフォンやタブレット端末などの軽薄短小なデバイスを想定しており, カメラと同じような感覚で, いつでもどこでも使うことができる.

In this paper, we propose a pants pattern design covering its highly precise design to pattern making by analyzing human body measurement data acquired with a 3D scanner. This system is characterized by: (1) being able to analyze data required for the design of pants patterns using an improved 3D human body measurement analysis application, and (2) being provided with automatic pants drafting functions installed in an apparel CAD system using 3D human body measurement data. As a result, individual 3D human body measurement data can be analyzed both accurately and rapidly, thereby enabling pants patterns that are comfortable and properly fit the physique of the wearer to be obtained in a short period of time.

Due to the recent development of underlying hardware technology and improvement in installing environments, public display has been becoming more common and attracting more attention as a new type of signage. Any signage is required to make its content more attractive to its viewers by evaluating the current attractiveness on the fly, in order to deliver the message from the sender more effectively. However, most previous methods for public display require time to reflect the viewers' evaluations. In this paper, we present a novel system, called Mood-Learning Public Display, which automatically adapts its content design. This system utilizes viewers' involuntary behaviors as a sign of evaluation to make the content design more adapted to local viewers' tastes evolutionarily on site. The system removes the current gap between viewers' expectations and the content actually displayed on the display, and makes efficient mutual transmission of information between the cyberworld and the reality.