Tokiichiro Takahashi

Bio: Tokiichiro Takahashi is an academic researcher from Tokyo Denki University. The author has contributed to research in topics: Rendering (computer graphics) & Tracing. The author has an hindex of 4, co-authored 48 publications receiving 778 citations.

Comprehensible rendering of 3-D shapes

Abstract: We propose a new rendering technique that produces 3-D images with enhanced visual comprehensibility. Shape features can be readily understood if certain geometric properties are enhanced. To achieve this, we develop drawing algorithms for discontinuities, edges, contour lines, and curved hatching. All of them are realized with 2-D image processing operations instead of line tracking processes, so that they can be efficiently combined with conventional surface rendering algorithms.Data about the geometric properties of the surfaces are preserved as Geometric Buffers (G-buffers). Each G-buffer contains one geometric property such as the depth or the normal vector of each pixel. By using G-buffers as intermediate results, artificial enhancement processes are separated from geometric processes (projection and hidden surface removal) and physical processes (shading and texture mapping), and performed as postprocesses. This permits a user to rapidly examine various combinations of enhancement techniques without excessive recomputation, and easily obtain the most comprehensible image.Our method can be widely applied for various purposes. Several of these, edge enhancement, line drawing illustrations, topographical maps, medical imaging, and surface analysis, are presented in this paper.

Non-photorealistic rendering technique for depicting 3D objects motion

Abstract: A still image extracted from a 3D animation sequence does not contain any information about the motions of 3D objects. In order to represent fast-moving 3D objects in a still image, the motion blur technique has been used. However, it cannot easily depict other kinds of motions. Several non-photorealistic rendering approaches [1] [2] have been proposed to depict the motions of 3D objects using “speed lines”. Speed lines, introduced by Masuch [1], are defined and drawn as difference vectors between the current and previous position vectors of the vertices of 3D object contours on a projected screen plane. Kawagishi et al. [2] generated very thin polygons instead of difference vectors. However, it is a tedious task to describe speed line types suitable for motion speed manually and repeatedly. We are studying a new non-photorealistic rendering technique that can depict various kinds of motions of 3D objects in animation scenes automatically and in real time. First, a motion is divided into a translation of and a rotation around the object’s center of gravity. For translation motion, we obtain the difference vectors of only the object’s center of gravity instead of those of the vertices of the 3D object’s contours. Texture-mapped polygons are generated as speed lines. Experimental results verify that our technique is effective and efficient enough to depict various motions of 3D objects automatically in real time.

Real-time Manga-like depiction based on interpretation of bodily movements by using Kinect

Abstract: In order to enrich visual communications, we propose a real-time system that interprets bodily movements by using Kinect [Kinect 2012]. Spatial and temporal positions of joints of body movements are captured by using Kinect. Based on captured data analysis, intensions of movements are interpreted. Finally, our system can automatically attach various Manga-like effects, such as speed lines, focus lines, or motion blur, to live action movies in real time. In addition, voice recognition function extends universality of our system's communication capabilities.

[Paper] A Method for Generating Pointillism Based on Seurat's Color Theory

Abstract: Georges-Pierre Seurat (1859-1891) is a pioneer, and prime exponent of \"neo-impressionism\". He established a painting technique called \"pointillism\" in which small and distinct dots of pure colors are applied in order to form a picture, instead of long strokes. One of the most important features of pointillism in neo-impressionism is based on the color theory such as optical mixture, complementary color contrast, and halo effect. Seurat applied the color theory to color arrangements particularly. Fig.1 (a) is the most famous work of Seurat, entitled \"Sunday afternoon on the island of La Grande Jatte\". According to the analysis on Seurat's work1)2), important characteristics of Seurat's pointillism are appeared in this work, and summarized as follows. (1) Pointillist painting: As shown in Fig.1 (a), pointillism is formed by small and distinct dots of pure colors instead of long strokes. (2) Use of primitive colors: 11 primitive color pigments from tubes are used without mixing them on his palette (Fig.2) for avoiding the reduction of saturation. (3) Control of luminance by white pigment: Only white pigment is mixed to each color on his pallet in order to control the luminance as shown in Fig.2. (4) Optical mixture is a visual effect of color fusion. Fig.1 (b) is an example of optical mixture. Juxtaposed vivid color dots cause similar spectra of colors Abstract We propose a method for generating pointillistic style images from input image considering the features of Seurat's pointillism. Georges-Pierre Seurat is a pioneer and prime exponent of neo-impressionist. He established a technique called pointillism based on scientific color theory. There are three important features of Seurat's pointillism: optical mixture, complementary color contrast and halo effect. The most important feature is the optical mixture. To implement the optical mixture faithfully, we present a pointillistic halftoning method for color halftoning on random dots by utilizing a spatial data structure of boundary sampling algorithm. In addition, we implement complementary color contrast and halo effect according to actual Seurat's painting steps.

An expression model of blood drying process based on SPH method

Abstract: Expression related to blood such as bloodshed and "blood splashes" is often used in "Computer Graphics field like game and movie". With the improvement of the performance of machines and technics, we able to express blood adhering to the object, however, there is few examples expressing the change of blood states over time. In this paper, we mainly focus on the coffee ring effect seen when blood is dried and we propose a method to simulate changes in the blood states like coffee ring effect, discoloration, over time and aim for more realistic representation of blood using the SPH method.

Playing for Data: Ground Truth from Computer Games

Abstract: Recent progress in computer vision has been driven by high-capacity models trained on large datasets. Unfortunately, creating large datasets with pixel-level labels has been extremely costly due to the amount of human effort required. In this paper, we present an approach to rapidly creating pixel-accurate semantic label maps for images extracted from modern computer games. Although the source code and the internal operation of commercial games are inaccessible, we show that associations between image patches can be reconstructed from the communication between the game and the graphics hardware. This enables rapid propagation of semantic labels within and across images synthesized by the game, with no access to the source code or the content. We validate the presented approach by producing dense pixel-level semantic annotations for 25 thousand images synthesized by a photorealistic open-world computer game. Experiments on semantic segmentation datasets show that using the acquired data to supplement real-world images significantly increases accuracy and that the acquired data enables reducing the amount of hand-labeled real-world data: models trained with game data and just \(\tfrac{1}{3}\) of the CamVid training set outperform models trained on the complete CamVid training set.

Learning hatching for pen-and-ink illustration of surfaces

Abstract: This article presents an algorithm for learning hatching styles from line drawings. An artist draws a single hatching illustration of a 3D object. Her strokes are analyzed to extract the following per-pixel properties: hatching level (hatching, cross-hatching, or no strokes), stroke orientation, spacing, intensity, length, and thickness. A mapping is learned from input geometric, contextual, and shading features of the 3D object to these hatching properties, using classification, regression, and clustering techniques. Then, a new illustration can be generated in the artist's style, as follows. First, given a new view of a 3D object, the learned mapping is applied to synthesize target stroke properties for each pixel. A new illustration is then generated by synthesizing hatching strokes according to the target properties.

Suggestive contours for conveying shape

Abstract: In this paper, we describe a non-photorealistic rendering system that conveys shape using lines. We go beyond contours and creases by developing a new type of line to draw: the suggestive contour. Suggestive contours are lines drawn on clearly visible parts of the surface, where a true contour would first appear with a minimal change in viewpoint. We provide two methods for calculating suggestive contours, including an algorithm that finds the zero crossings of the radial curvature. We show that suggestive contours can be drawn consistently with true contours, because they anticipate and extend them. We present a variety of results, arguing that these images convey shape more effectively than contour alone.

Illustrating smooth surfaces

Abstract: We present a new set of algorithms for line-art rendering of smooth surfaces. We introduce an efficient, deterministic algorithm for finding silhouettes based on geometric duality, and an algorithm for segmenting the silhouette curves into smooth parts with constant visibility. These methods can be used to find all silhouettes in real time in software. We present an automatic method for generating hatch marks in order to convey surface shape. We demonstrate these algorithms with a drawing style inspired by A Topological Picturebook by G. Francis.

Computer-generated watercolor

Abstract: A watercolor model based on an ordered set of translucent glazes, which are created independently usinig a shallow water fluid simulation. A Kubelka-Munk compositing model is used for simulating the optical effect of the superimposed glazes. The computer generated watercolor model is used as part of an interactive watercolor paint system, or as a method for automatic image “watercolorization.”