Details of a Researcher - Fujishiro, Issei

Papers - Fujishiro, Issei

Division display 221 - 240 of about 416 ／ All the affair displays >>

小林杏理，藤代一成

映像情報メディア学会技術報告 (映像情報メディア学会) 40 ( 11 ) 261 - 264 2016.03

Research paper (conference, symposium, etc.), Joint Work

本稿では, 即興演奏を行う楽器演奏者を対象ユーザとして, その演奏情報をリアルタイムに読み取り, それに呼応して映像生成を行うシステムを提案した. 提案システムでは, 演奏者の演奏をアフェクティブな評価基準をもって評価し, その結果に応じて映像生成を行い, 演奏者にインスピレーションを与えることを目指す. インスピレーションを得た演奏者が演奏を変化させ, またシステムはそれに呼応し映像を変化させるという, アフェクティブな表現のループを形成することで, 演奏者の「共演者」としての役割をシステムで再現することが, 本研究の目標である.

マルチスレッド設計によるインタラクティブなグルーヴ可視化

神展彦，芳賀直樹，藤代一成

映像情報メディア学会技術報告 (映像情報メディア学会) 40 ( 11 ) 1 - 4 2016.03

Research paper (conference, symposium, etc.), Joint Work

グルーヴとは，演奏における音のうねり，リズムのノリや⼀体感などを表す, よい演奏に必要不可欠な要素のひとつである．グルーヴ可視化は，音楽的センスや聴力の個人差によらないグルーヴの習得や音学経験に制限されない音楽の感動の共有など，さまざまな応用につながる．本稿では，MIDI 信号として入力された音楽をインタラクティブに図形に変換する，直感的なリズム可視化システムSeeGroove2を提案した．演奏された音符それぞれに対して点をプロットし，滑らかに補間することにより，リズムパターンを多様な形に表現する．演奏に対してインタラクティブに図形が変化するので，ユーザは音楽の移り変わりを視覚的に感じることができる．このインタラクティブ性は，1,920Hzで動作する入力系スレッドと60Hzで動作する計算・描画系スレッドのマルチスレッド設計によって実現されている．

薄膜と液状内容物との双方向連成を考慮した破裂のビジュアルシミュレーション

佐原淳允，藤代一成

映像情報メディア学会技術報告 (映像情報メディア学会) 40 ( 11 ) 37 - 40 2016.03

Research paper (conference, symposium, etc.), Joint Work

風船やクッションなどの薄膜で覆われたオブジェクトの破裂表現は，日常頻繁に見られる現象であるが，正確にモデリングされた例はまだ報告されていない．そこで本研究では，破裂を『薄膜に拘束されていた液状内容物が，薄膜の破壊によって流出し，飛沫を発生させながら飛び散る』現象と定義し，モデリングを試みた．特に，内容物から薄膜への影響だけでなく，薄膜から内容物への影響も正確に表現する双方向のモデルの構築を目指す．具体的なシミュレーション手法として，薄膜を三角形要素，内容物を四面体要素で記述し，応力の計算に有限要素法を使用する．本研究のモデルでは，メッシュの境界面が陽に構築されているため，衝突や摩擦などの表面の相互作用の計算を厳密に定義することができる．また，各タイムステップの最後には，計算精度の向上や，もっともらしい破片生成のために要素のリメッシングを行う．提案手法では，薄膜の破壊が引き金となり，破裂が発生するという，既存の数値計算モデルでは表現できない現象をシミュレーションすることが可能であることを，実物の高速撮影ビデオとの定性的比較により明らかにした．

作曲家の変奏特徴を理解するための変奏曲可視化システム

村岡遼太，藤代一成

映像情報メディア学会技術報告 (映像情報メディア学会) 40 ( 11 ) 89 - 92 2016.03

Research paper (conference, symposium, etc.), Joint Work

変奏曲とは，主題とよばれる簡単な旋律と主題を様々に変化させた変奏からなるクラシック音楽の楽曲形式の一つである．変奏曲を理解する際には，主題がどのように変化しているかを把握することが重要であるが，楽器経験者でも楽譜や音源のみから主題の変化の様子を読み取ることは困難である．そこで本研究では，楽器の演奏者や，音楽は聴くが演奏経験はないといった音楽愛好家に対して，変奏曲の構造を可視化するシステムを開発した．本システムは，入力をMIDI データとし，3 次元空間にオブジェクトを配置することで可視化する．その際，用いられている変奏様式によりオブジェクトの色や形状を変化させることでどのような変奏がなされてかを表現している．また，ユーザがより変奏曲の構造を理解しやすくするよう対話的視覚情報探索機能を提供する．システムの性能を評価するにあたり，マッピングのわかりやすさ，楽譜と本システムを用いた際の変奏曲の理解度についての実験を行い，本システムが変奏曲の構造理解に有効であることが示された．

非一様な密集度を考慮した群衆移動のモデリング

鈴木涼平，藤代一成

映像情報メディア学会技術報告 (映像情報メディア学会) 40 ( 11 ) 127 - 130 2016.03

Research paper (conference, symposium, etc.), Joint Work

群衆の動きに対する研究は，古くは人間の認知に関する研究から始まり，今日に至るまで発展を続けてきている．群衆の動きを自動生成するモデルに対する需要は幅広く，娯楽志向のものに目を向けると，多様なシーンに対して，尤もらしい人の動きを高速に計算できるモデルが求められている．本稿では，リアルタイム性が必要となる状況下で，人の動きに関してユーザが満足できる品質を達成するため，カット割りができない状況でも，シーンを占める群衆の密集度が大幅に変化しても，頑健かつ高速な動作を保証できる，群衆移動のモデルを提案した．このモデルでは，個人としての人の動きだけではなく，近しい関係の人によって形成される，2～3人の小集団の動きも扱うことで，より尤もらしい群衆移動を自動生成する．

An object space anti-aliasing method for shadows of hair-shaped objects

IKEDA Yasunari, MATSUOKA Toru, FUJISHIRO Issei

The IEICE Transactions on Information and Systems (IEICE) J99-D ( 3 ) 307 - 316 2016.03

Research paper (scientific journal), Joint Work, Accepted, ISSN 1881-0225

In this paper, we focus our attention to the shadow of hairs, which are vital to human animation, but are not easy to be accompanied by their stable shadows. Temporal-aliasing artifacts could be alleviated by image-space post-processing such as neighboring pixel averaging, though the quality depends strongly on the density of projected pixels. We therefore proposed an object-space anti-aliasing method by approximating the 3D shape of hair with a cloud of variable-size square splats with concentric opacity distribution. Empirical evaluation revealed that the proposed method outperforms the existing image-space
anti-aliasing with minimum loss of computational speed.

Access to Document (DOI)

Volume preserving viscoelastic fluids with large deformations using position-based velocity corrections

TETSUYA Takahashi, DOBASHI Yoshinori, FUJISHIRO Issei, NISHITA Tomoyuki

The Visual Computer (Springer-Verlag) 32 ( 1 ) 57 - 66 2016.01

Research paper (scientific journal), Joint Work, Accepted, ISSN 0178-2789

We proposed a particle-based hybrid method for simulating volume preserving viscoelastic fluids with large deformations. Our method combines smoothed particle hydrodynamics (SPH) and position-based dynamics (PBD) to approximate the dynamics of viscoelastic fluids. While preserving their volumes using SPH, we exploit an idea of PBD and correct particle velocities for viscoelastic effects not to negatively affect volume preservation of materials. To correct particle velocities and simulate viscoelastic fluids, we use connections between particles which are adaptively generated and deleted based on the positional relations of the particles. Additionally, we weaken the effect of velocity corrections to address plastic deformations of materials. For one-way and two-way fluid-solid coupling, we incorporate solid boundary particles into our algorithm. Several examples demonstrated that our hybrid method can sufficiently preserve fluid volumes and robustly and plausibly generate a variety of viscoelastic behaviors, such as splitting and merging, large deformations, and Barus effect.

Access to Document (DOI)

Development of a pants pattern design system based on 3D body data analysis

YAMAMOTO Takami, NAKAYAMA Masanori, KATSURA Rui, SAKAMOTO Akira, FUJISHIRO Issei

Transactions on Image Electronics and Visual Computing (IIEEJ) 3 ( 2 ) 114 - 124 2015.12

Research paper (scientific journal), Joint Work, Accepted

In this paper, we proposed a pants pattern design system covering its highly precise application 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 comfortably and properly fit the physique of the wearer to be obtained in a short period of time.

Ontology-supported design of visualization workflows in fluid science

TAKESHIMA Yuriko, FUJISHIRO Issei

Proceedings of ACM SIGGRAPH Asia 2015 Visualization in High Performance Computing (ACM, New York, NY, USA) 2015.11

Research paper (international conference proceedings), Joint Work, Accepted

Modular visualization environments (MVEs) are commonly used in many disciplines including fluid science. However, it is difficult for scientists and engineers, who are not necessarily visualization experts, to make full use of many built-in modules on the basis of their usability and interdependency for constructing optimal visualization workflows. In this paper, we presented a novel MVE, called goal-oriented application design guidance for modular visualization environments/flow visualization (GADGET/FV), that is designed and built on an ontological framework to assist users to design their flow visualization workflows. The system relies primarily on an extended Wehrend Matrix that allows users to specify their goals and requirements, and to search the knowledge-base for appropriate visualization workflows on their behalf. In addition, users can browse successful examples maintained in a case repository to facilitate their workflow design further; this concept is referred to as "design by example." In this paper, examples of three practical fluid analyses were presented to illustrate the usefulness of a prototype system. Furthermore, the extensibility of the system to handle large-scale datasets was discussed.

Access to Document (DOI)

Adsorptive SPH for directable bleeding simulation

UEDa Kazuhide, FUJISHIRO Issei

Proceedings of the 14th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry (VRCAI 2015) (ACM, New York, NY, USA) 9 - 16 2015.10

Research paper (international conference proceedings), Joint Work, Accepted

We built upon SPH to devise a directable scheme for bleeding simulation that realistically represents adherence onto and streak formation on the skin surface in real time. To ensure compatibility of expressiveness and space & time efficiency, we judiciously designed four specific types of particles and their interactions and incorporated these elements into the SPH scheme. In-vivo particles obey a simple in-vivo blood flow model that can reflect both hemorrhage volume and blood pressure. The particles flowing out through the wound, called flow particles, manifest the primary behaviors of bleeding, including blood coagulation, blood adherence onto the skin surface, and streak formation on the skin surface. The scheme then involves transforming the flow particles adhered onto the skin surface into relatively small adhered particles and representing a thin, flat layer of adhered blood with anisotropic kernels. Guiding particles generated along user-input strokes play a distinct role in giving gravity, biased slightly by the guiding force, to the remaining flow particles and thereby allow the users to control the overall bleeding so that it follows the desirable traces.

Access to Document (DOI)

SeeGroove: Supporting groove learning through visualization

FUJISHIRO Issei, HAGA Naoki, NAKAYAMA Masanori

Proceedings of 2015 International Conference on Cyberworlds (IEEE) 189 - 192 2015.10

Research paper (international conference proceedings), Joint Work, Accepted

Groove is a global feature of music performance that generalizes the surges, propulsive rhythmic feel, dynamics, and togetherness of sound to reflect overall rhythmic pleasure. Groove is vital to a good performance, though it is claimed that consierable experience is needed to be accumulated for the expression of groove at will. In this work, we proposed a general visualization system, called SeeGroove, that allows performers to efficiently comprehend groove in various music genres. For each genre, the standard MIDI dataset was extracted from actual performances of professional players in advance. The system then takes the difference between an input data of performance and the standard of a chosen genre to quantify the degree of groove of the performance. To visualize the differences, the user is allowed to rely on any combination of two modes and two views of designated visualization in accordance with his/her preferences and target information.

Access to Document (DOI)

IFWB: A simple image-based modeling approach to breaking the “fourth wall”

ISAKA Takeru, FUJISHIRO Issei

The Journal of the Society for Art and Science (The Society for Art and Science) 14 ( 4 ) 150 - 159 2015.09

Research paper (scientific journal), Joint Work, Accepted, ISSN 1347-2267

This paper proposed 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, and then IFWB augments its depth without any further input and allows him/her to walk through the world of image freely. 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.

Axis contraction of parallel coordinates using spectral graph analysis

NOHNO Koto, WU Hsiang-Yun, WAYANABE Kazuho, TAKAHASHI Shigeo, FUJISHIRO Issei

Journal of IIEEJ (IIEEJ) 44 ( 3 ) 447 - 456 2015.08

Research paper (scientific journal), Joint Work, Accepted, ISSN 0285-9831

Parallel coordinates is well-known as a popular tool for visualizing the underlying relationships among variables in high-dimensional datasets. This visualization technique is useful for visually understanding the degree of correlation between data samples in terms of two adjacent axes. However, this representation still suffers from distracting visual clutter especially when the numbers of data samples and their associated dimension become high, because the associated polyline samples intricately overlap with each other within the limited screen space. This paper presented a method of alleviating such visual clutter by contracting multiple axes through the analysis of correlation between every pair of variables. In this method, we first define the similarity between a pair of dimensions as the value of the correlation coefficient, and construct a subgraph from the complete graph through eliminating all the edges in that their absolute correlation coefficients are less than some threshold, and then reorder the multiple axes by projecting the nodes onto the primary axis obtained using the spectral graph analysis. This allows us to compose a dendrogram tree by recursively merging a pair of the closest axes one by one. Smooth animation of the
associated axis contraction and expansion has also been implemented to enhance the visual readability of behavior inherent in the given high-dimensional datasets. We also conducted a user study to investigate how the correlations among coordinate axes are better visualized using our approach.

Glasses-free 3D imaging by optic illusion with orthogonally-arranged multi-displays

井阪　建，藤代一成

Visual Computing/グラフィクスとCAD合同シンポジウム2015 DVD予稿集 (画像電子学会) 2015.06

Research paper (conference, symposium, etc.), Joint Work, Accepted

Motion puppetry taking skeletal similarity into account

KANEKO Norihide, TAKAHASHI Reo, FUJISHIRO Issei

Proceedings of the 28th International Conference on Computer Animation and Social Agents (CASA2015) (Computer Graphics Society) 2015.05

Research paper (international conference proceedings), Joint Work, Accepted

Motion retargeting is a well-known method for generating character animations in computer graphics, where the original motion of an actor can be transferred to an avatar with a different skeletal structure and/or limbs' length. A specific type of retargeting for manipulating various avatars interactively is referred to as puppetry. However, previous learning-based works of puppetry cannot be said to be fully interactive, or require a large amount of metadata about the avatar's skeleton as well as training data of motions. Thus, we attempted to integrate existing methods by taking into account skeletal similarity between an actor and his/her avatar, to come up with an interactive and intuitive motion retargeting method that only requires a relatively small amount of training animations together with simple input metadata. Moreover, by classifying avatar's body parts in a procedural manner, a more flexible motion puppetry was realized, where the system user is allowed to specify desirable part-to-part correspondences.

Implicit formulation for SPH-based viscous fluids

TAKAHASHI Tetsuya, DOBASHI Yoshinori, FUJISHIRO Issei, NISHITA Tomoyuki, LIN Ming C.

Computer Graphics Forum (Eurographics) 34 ( 2 ) 493 - 502 2015.05

Research paper (scientific journal), Joint Work, Accepted, ISSN 1467-8659

We proposed a stable and efficient particle-based method for simulating highly viscous fluids that can generate coiling and buckling phenomena and handle variable viscosity. In contrast to previous methods that use explicit integration, our method uses an implicit formulation to improve the robustness of viscosity integration, therefore enabling use of larger time steps and higher viscosities. We use Smoothed Particle Hydrodynamics to solve the full form of viscosity, constructing a sparse linear system with a symmetric positive definite matrix, while exploiting the variational principle that automatically enforces the boundary condition on free surfaces. We also proposed a new method for extracting coefficients of the matrix contributed by second-ring neighbor particles to efficiently solve the linear system using a conjugate gradient solver. Several examples demonstrated the robustness and efficiency of our implicit formulation over previous methods and illustrated the versatility of our method.

Access to Document (DOI)

Biclustering multivariate data for correlated subspace mining

WATANABE Kazuho, WU Hsiang-Yun, NIIBE Yusuke, TAKAHASHI Shigeo, FUJISHIRO Issei

Proceedings of IEEE Pacific Visualization Symposium 2015 (IEEE Computer Society) 287 - 294 2015.04

Research paper (international conference proceedings), Joint Work, Accepted, ISSN 2165-8765

Exploring feature subspaces is one of promising approaches to analyzing and understanding the important patterns in multivariate data. If relying too much on effective enhancements in manual interventions, the associated results depend heavily on the knowledge and skills of users performing the data analysis. This paper presented a novel approach to extracting feature subspaces from multivariate data by incorporating biclustering techniques. The approach has been maximally automated in the sense that highly-correlated dimensions are automatically grouped to form subspaces, which effectively supports further exploration of them. A key idea behind our approach lies in a new mathematical formulation of asymmetric biclustering, by combining spherical k-means clustering for grouping highly-correlated dimensions, together with ordinary kmeans clustering for identifying subsets of data samples. Lower dimensional representations of data in feature subspaces are successfully visualized by parallel coordinate plot, where we project the data samples of correlated dimensions to one composite axis through dimensionality reduction schemes. Several experimental results of our data analysis together with discussions were provided to assess the capability of our approach.

Access to Document (DOI)

Development of a Skirt Pattern Design System Based on 3D Human Body Measurement Data Analysis

YAMAMOTO Takami, NAKAYAMA Masanori, KATSURA Rui, SAKAMOTO Akira, FUJISHIRO Issei

The Journal of the Society for Art and Science (The Society for Art and Science) 14 ( 1 ) 9 - 19 2015.03

Research paper (scientific journal), Joint Work, Accepted, ISSN 1347-2267

This paper presents a design system encompassing the process from high-precision design to the production of a skirt prototype based on the deployment of design by analysis of 3D body data gathered with a 3D scanner. This system is characterized in that (1) it allows the generation of accurate reference cross-sectional data from polygonal data of the body according to an intended design, (2) automatic drawing functions by the draping method using a reference cross-sectional data, and (3) production of patterns capable of accommodating various body types , As a result, individual 3D body data analysis can be performed rapidly and wear comfortable skirt patterns that fit the body contours can be obtained in a shorter period of time.

骨格の類似度を考慮したモーションパペトリー

KANEKO Norihide,TAKAHASHI Reo,FUJISHIRO Issei

映像表現・芸術科学フォーラム2015 (映像情報メディア学会・画像電子学会・芸術科学会) 2015.03

Joint Work

SeeGroove：可視化を介したグルーヴの学習支援

HAGA Naoki,NAKAYAMA Masanori,FUJISHIRO Issei

映像表現・芸術科学フォーラム2015 (映像情報メディア学会・画像電子学会・芸術科学会) 2015.03

Joint Work

　 Previous page - Next page 　