SIGGRAPH ASIA 2018 | Poster | 2018.12

Pre- and post-processes for automatic colorization using a fully convolutional network

Sophie Denise, Paulette Ramassamy, Hiroyuki Kubo, Takuya Funatomi, Daichi Ishii, Satoshi Nakamura, Yasuhiro Mukaigawa, Akinobu Maejima

Automatic colorization is a significant task especially for Anime industry. An original trace image to be colorized contains not only outlines but also boundary contour lines of shadows and highlight areas. Unfortunately, these lines tend to decrease the consistency among all images. Thus, this paper provides a method for a cleaning pre-process of anime dataset to improve the prediction quality of a fully convolutional network, and a refinement post-process to enhance the output of the network.
Proceeding SA '18 SIGGRAPH Asia 2018 Posters Article No. 70

SIGGRAPH ASIA 2018 | Technical Brief: Data Acquisition | 2018.12

Creating a Virtual Human that Visualizes Skin Strain Distribution for Apparel Wearing Simulation

Akihiro Ozawa, Ken Kusano, Hiroto Mori, Takehiro Tagawa, Yusuke Tonuma, Yoshiyuki Kubo, Akinobu Maejima, Tatsuo Yotsukura

We introduced the first step of our research collaboration to create a virtual human that visualizes skin strain distribution to aid the design of comfortable, ergonomic sportswear.
Proceeding SA '18 SIGGRAPH Asia 2018 Technical Briefs Article No. 33

SIGGRAPH ASIA 2018 | Poster: Animation and Visual Effects | 2018.12

Blur Algorithms for Cartoon Animation

Marc Salvati, Kota Ito

This talk introduces specific blurring algorithms tailored for cartoon animation: a "Soft" style blur, and new directional/radial anisotropic blur used for cartoon moody and motion lines.
Proceeding SA '18 SIGGRAPH Asia 2018 Posters Article No. 3

CGWORLD 2015 Creative Conference | 2015.11.22

毛、生やしています – Furワークフロー構築 –

Yoshinori Moriizumi, Tatsuo Yotsukura


SIGGRAPH ASIA 2015 | Workshop: R&D in the Video Game Industry | 2015.11.2

Shader Development At OLM

Shinji Ogaki

Recently our studio started using Arnold renderer to create cinematics for games. It was not an easy task for artists to optimize shader parameters for a modern physically-based renderer because they were not familiar with it. In this paper we describe the challenges we had and how our shaders were developed to tackle this situation.
Proceeding SA '15 SIGGRAPH Asia 2015 R&D in the Video Game Industry Article No. 5

SIGGRAPH ASIA 2015 | Technical Brief: Animation | 2015.11.2

A Linear Blending Scheme for Rigid and Non-Rigid Deformations

Liu Gengdai, Ken Anjyo

We present a novel linear blending scheme utilizing existing barycentric coordinates. The scheme enables cage vertices associated with sparse weights to be inferred automatically by a user's manipulation on constrained vertices or handles in real-time. Our prototype system demonstrates that our linear blending scheme can deal with rigid and non-rigid deformation more consistently and efficiently than previous approaches.
Proceeding SA '15 SIGGRAPH Asia 2015 Technical Briefs Article No. 19

CGWORLD 2014 Creative Conference | 2014.11.23

ド根性映像制作脱出までの険しい道のり – 相棒 (R&Dアーティスト) 募集中

Marc Salvati


DigiPro 2014 | 2014.8.9

The Brush Shader: A Step towards Hand-Painted Style Background in CG

Marc Salvati, Ernesto Ruiz Velasco, Katsumi Takao

We will detail how we reproduced hand-painted texture and background by simulating Photoshop functionalities in a shader: the Brush Shader. This results in much detailed shading, and releases us from restrictions on camera work related to matte painting imperatives.

CGWORLD 2013 Creative Conference | 2013.10.27

ド根性映像制作をやめたい!どうやって? – 快適映像制作のためのド根性ツール開発

Marc Salvati

2012年の「ド根性映像制作をやめないか? 映像制作を支える技術の大切さ」では、ド根性が求められる映像制作現場の内情、研究開発部門の役割、また技術の大切さについて発表しました. 今回はさらに踏みこんで、具体的に技術の開発事例をとりあげます.多忙を極め、技術開発に費やす時間のないド根性映像制作のなかで、技術をつくるにはどうすればよいのか?を考えます.すこしでもド根性映像制作の世界を変えるきっかけになれば幸いです.


SIGGRAPH 2013 Computer Animation Festival - Production Sessions

OLM Digital Presents the Anime Spirit: From Pokémon and Pac-Man to Live Action Films

Yasuhiro Mikami, Masashi Kobayashi, Moto Sakakibara, Ken Anjyo

Anime has gained great popularity in the world for its unique expressiveness in contrast to western animation. OLM Digital, a digital production company in Tokyo, keeps trying new anime styles, making the Pokémon movies over 15 years. This session presents the company’s various works in 2D/3D hybrid anime, 3DCG, S3D anime and live action films. The showcase focuses on how the anime spirit of OLM Digital is put into various visual forms. The brand-new Pac-Man animated TV series, which is a collaborative work with Sprite Animation Studios, is also one of the highlights of this session.

DigiPro 2013 | 2013.7.20

MorphoLogical Anti-Aliasing in Production

Marc Salvati, Yosuke Katsura, Ken Anjyo

Morphological Anti-Aliasing (MLAA) has been successfully applied in the game industry as a cheap and efficient method to alleviate the cost of Multi-Sampling Anti-Aliasing (MSAA) but also has been used in cel animation production for years. In this presentation we will give details about how we use MLAA and about our specific implementation of the algorithm.

CGWORLD 2012 Creative Conference | 2012.10.28

ド根性映像制作をやめないか?- 映像制作を支える技術の大切さ

Marc Salvati

日本では研究開発部門をもっていたり、技術のわかるテクニカルアーティストやエンジニアを雇っている映像プロダクションはまだ少ないのが現状です.厳しい予算やスケジュールのなかで、実際どう乗り越えているかというと、スタッフの”根性”で何とかしていたりします.しかし、どんどん高度になるCG技術に対して、今後も根性で太刀打ちできるでしょうか? 高度な技術を理解し、応用、発展させて実際に制作環境で使えるようにできるテクニカルスキルが今後さらに必要になっていくでしょう.講演ではOLMデジタルで実際にどのように技術を活用しているかを紹介します.システムの管理からパイプライン、自動化ツール、そして新しい表現を可能にする研究開発、技術教育まで幅広く取り上げます.技術の壁を越えれば日本の映像制作業界はもっとパワーアップできると思います.


Digital Production Symposium (DigiPro) 2012 | 2012.8

Extracting Fluid from a Video for Efficient Post-Production

Makoto Okabe, Ken Anjyo, Rikio Onai

We propose a method to extract fluids from a video that is captured outside a special studio. The method allows the designer to efficiently use the extracted fluids for the post-production.

SIGGRAPH ASIA 2011 Poster, Production & Visual Effects | 2011.12

Hybrid Framework for Blendshape Manipulations

Hideki Todo, Ken Anjyo

We propose a hybrid blendshape manipulation framework to accelerate the face posing process. The hybrid framework enables interactive edits with direct manipulation and face-constraint-relaxation, which also avoid face geometry distortion during operations. An underlying primitive in this framework is a constraint blendshape model, which operates indirectly sliders by solving a simple least square problem. We demonstrate the effectiveness of our framework with a few application results.

SIGGRAPH 2011 Talk, Session: Changing Dimension | 2011.8

Developing Tools for 2D/3D Conversion of Japanese Animations

Marc Salvati, Yosuke Katsura, Tatsuo Yotsukura, Miki Kinoshita, Ken Anjyo, Hiroshi Uchibori

Production of “Inazuma Eleven: the Movie” required a set of new tools to enable conversion from Japanese animation to 3D stereoscopic movie. This talk summarizes those in-house tools and how OLM used Japanese animation specificity to build up an efficient conversion system.

SIGGRAPH ASIA 2009 Poster, Production | 2009.12

Asset management system for digital production workflow

Tatsuo Yotsukura, Miki Kinoshita, Satoru Yamagishi, Kazuyuki Ishihara, Yoshinori Moriizumi

This paper describes the asset management system (AMS) that enables sharing, manipulation and version management of various digital data in production workflow at OLM Digital. In the past, our digital asset has been managed by each film work manager. Each manager uses his own file and directory naming convention. Unfortunately, these rules were sometimes ambiguous and we had various problems like file loss, duplications and version conflicts. It was also difficult to find and use data from previous projects.

SIGGRAPH ASIA 2009 Sketch, Production Session 2: Research & Technique | 2009.12

Identifying salient points

J. P. Lewis, Ken Anjyo

The definition of “important” or salient points on a shape is an old and fundamental problem in computer graphics. Important points, which we will term key points, can be used for representing a shape (e.g. as vertices of a polyline or as knots of a spline). Key points that correspond to perceptually salient points are useful as handles for editing. Key points are also good points to retain in simplifying a shape.

SIGGRAPH ASIA 2009 sketch, Production Session 1: Pipeline | 2009.12

Rich Background for Cel Animation

Yosuke Katsura, Ken Anjyo

This sketch demonstrates seamless merger of cel animation and 3D CG scenes for a huge amount of animated objects.

SIGGRAPH 2008 Talk, Smile for the Camera | 2008.8

Extracting higher-level information from facial mocap

J. P. Lewis, Ken Anjyo

Motion capture data is valuable information, but in a low-level form that makes subsequent editing operations difficult. In cases where extensive manipulation is needed a manually animated performance maybe be preferable, because the manually constructed representation has “artist-friendly” higher level controls. Algorithmic approaches can potentially extract higher-level and semantic representations from motion capture, but much work remains to be done in this area. In this talk we show examples of several types of higher level information that can be extracted using an affinity measure on the mocap data.

SIGGRAPH 2007 Poster, Animation | 2007.8

Directable Crowd Animation

Wataru Ishihara, Ken Anjyo

We present an interactive system for creating directable crowd animation, focusing on making the CG crowd shots suited especially for a TV program. Our system cannot make such “advanced” crowd animations as those for a blockbuster movie, which were made, for example, using AI techniques. However, our system allows a user to quickly make the crowd scenes under a limited time and budget. Our “handy” crowd system is currently available as a Maya plug-in our workplace, offering several features, which we believe would be useful enough to quickly make various crowd shots.

SIGGRAPH 2007 sketch, Looking Good | 2007.8

Anime Perspective

Yosuke Katsura, Ken Anjyo

In traditional hand-drawn animation, the perspective view is not geometrically correct, unlike 3DCG. However, anime perspective is more natural for human eyes. We present two anime perspective projection methods for seamlessly merging 3D models and traditional 2D animation.

SIGGRAPH 2004 | Sketch, Sketchy Sketches

Tweaking stylized light and shade

Ken Anjyo, Shuhei Wemler

Interactive techniques for creating fake, stylized highlights and shaded areas simply by drag operations on an on-screen surface, without laborious parameter tuning.
ACM SIGGRAPH 2004 Sketches (SIGGRAPH ’04) Page 5 , 2004.8

SIGGRAPH 2004 | Sketch, Art on the Small Screen

Workflow and CG Tools for the Cartoon TV Program “Monkey Turn”

Satoru Yamagishi, Megumi Kondo, Hiroshi Uchibori, Ken Anjyo

“Monkey Turn” is a cartoon TV program in Japan weekly-aired since January 2004. The highlights of this cartoon animation are dynamic motorboat race scenes, which are achieved by hybrid use of 3D CG and hand-drawn objects. This sketch describes the workflow, CG software plug-ins and script tools specialized for efficiently making the animation program under restricted time and work force.
ACM SIGGRAPH 2004 Sketches (SIGGRAPH ’04) Page 105 , 2004.8

SIGGRAPH 2001 | Sketch, Merging 2D and 3D Production Methods

Digital Tricks for Japanese Cel Animation

Akio Kazumi, Megumi Kondo, Katsuaki Hiramitsu, Ken Anjyo, Yoshinori Dobashi

How digital tricks are created and applied in making Japanese cel animation, including practical examples from the Pokémon movies..
Our topics in this talk are:
- Difference between Western and Japan in production style
- Techniques of merging 2D and 3D for Japanese cel animation
ACM SIGGRAPH2001 Animation & Video Sketches, 2001.8


SIGGRAPH 2014 Course | 2014.8.13

Mathematical Basics of Motion and Deformation in Computer Graphics

Hiroyuki Ochiai, Ken Anjyo

The objective of this course is to fill the gap between the original mathematical concepts and the practical meanings in computer graphics without assuming any prior knowledge of pure mathematics. It illustrates the practical utility of deep understanding of the mathematical concepts.
Proceeding SIGGRAPH '14 ACM SIGGRAPH 2014 Courses Article No. 19

SIGGRAPH 2014 Course | 2014.8.14

Scattered Data Interpolation for Computer Graphics

Ken Anjyo, J.P. Lewis, Frédéric Pighin

The course presentation assumes knowledge of linear algebra, as will be familiar to computer graphics programmers and technical directors. We have developed novel explanations of concepts such as Greens functions using only linear algebra concepts. One appendix introduces a functional analysis view of interpolation and requires calculus.
Proceeding SIGGRAPH '14 ACM SIGGRAPH 2014 Courses Article No. 27


SIGGRAPH 2011 Course | 2011.8

Stereoscopy, from XY to Z

Samuel Gateau, Robert Neuman, Marc Salvati

This comprehensive course summarizes standard stereo-projection techniques, audience depth perception and comfort factors, recommended uses of depth as a storytelling tool, how stereo is used in gaming and real-time applications, details on conversion of 2D content to 3D, guidance on new creative choices, and using depth in an aesthetic way.


SIGGRAPH ASIA 2009 Course | 2009.12

Crowd Animation: Tools, Techniques, and Production Examples

Craig “Xray” Halpern, Ken Anjyo, Mihai Cioroba, Paul Kanyuk, Stephen Regelous, Takashi Yoshida, Marc Salvati

By examining the use of crowd animation across several production pipelines at different studios, this course reveals various methods and solutions for animating, simulating, and rendering crowd animation. It presents an overview of the history and concepts of crowd animation, a review of the current state of the art in crowd animation, and some thoughts on the future of this growing field of computer graphics.

©Nintendo·Creatures·GAME FREAK·TV Tokyo·ShoPro·JR Kikaku ©Pokémon ©2012PIKACHU PROJECT ©Katsutoshi Kawai·Shogakukan/Monkey Turn Project·TV Tokyo ©L5/FCIE M 2010 2014 ©Pokémon ©1995-2014 Nintendo / Creatures Inc / GAME FREAK Inc ©BANDAI/TV TOKYO·ここたま製作委員会 ©LEVEL-5/FC Inazuma eleven·TV TOKYO ©LMYWP2016