KEYNOTE: Large-scale physics-based simulations and scientific visualizations
Professor Takayuki Aoki | Tokyo Institute of Technology, Japan
Real-world simulations based on physics require huge amount of computer resources. It is one of promising ways to use a GPU as an accelerator for scientific computing. We demonstrate several large-scale applications carried out on the TSUBAME supercomputer equipped with 4,000 GPUs at the Tokyo Institute of Technology. An air flow simulation for 10 km x 10 km area of metropolitan Tokyo with 1-m resolution is computed on 52 billion meshes by the Lattice Boltzmann method, in which a new LES (Large-Eddy Simulation) model has been employed. We also show gas-liquid two-phase flows using the VOF method and a tsunami debris flow including a lot of floating materials with a SPH-DEM coupled solver. The large-size output data are rendered by ray tracing to make realistic visualizations.
Prof. Aoki has research interests in large-scale applications of computational fluid dynamics including weather prediction, two-phase flow, and fluid-structure interaction. He has been studying higher-order numerical schemes. His current research topics is the dynamic load balance for stencil and particle applications. He has been a pioneer in GPU computing since the beginning of GPGPU and recognized as a NVIDIA CUDA Fellow. He organizes a consortium of GPU computing with more than 1,000 members in Japan. He has received many awards highlighted by the MEXT Minister award, the Achievement Award from JSME, JSIAM and the ACM Gordon Bell Prize in 2011. He authored the first textbook on CUDA programming in Japanese. He is a fan of stereoscopic and 360 degree movies.
WebVR – Virtual Reality on the Web
Sandy Ressler | NIST USA
Abstract: Recent excitement about Virtual Reality appears to be bringing about another round of VR hype. In the mid 1990’s we saw VR explode only to fall flat. This time things look different. One key difference is the use of 360 degree videos and the combination of those videos with the ubiquitous nature of the Web. VR on YouTube using your cellphone for a display device is bringing immersive content to the masses. In this talk we take an entertaining look at all of these exploding technologies and some applications.
Link to presentation.
Sandy Ressler has nearly 40 years of experience with a wide variety of computer graphics and user interface technologies. Starting a career at Bell Labs followed by a stint at a video game startup he is a pioneer and visionary in the development of 3D computer graphics for use on the Internet. He was on the Web3D Consortium’s, Board of Directors for 6 years, two of which as Vice President. The consortium is the organization responsible for VRML/X3D the ISO standard for 3D on the Internet. He is currently exploring ways to integrate Virtual Reality with the Web at the National Institute of Standards and Technology where he get’s away with all this play disguised as work.
From 1997-2001, he created and ran the world’s leading web site for 3D on the web at About.com. Ressler ran several Web3D Showcase events (demonstration events) at SIGGRAPH (the premier conference for the computer graphics industry) which exposed tens of thousands of people to Web3D applications. Ressler has also been a leader in applying visualizations for 3D anthropometry (human measurement) data, resulting in safer children’s products, and was a leader in the use of Web3D graphics for manufacturing applications.
He has published over 20 peer-reviewed articles (www.sandyressler.com/vitae.htm) and is widely regarded as one of the leading figures in the industry as illustrated by articles in the press including the NY Times, Federal Computer Week, and Popular Mechanics. Ressler has also authored three books, two on electronic publishing and coauthored the classic “Life with UNIX”. He holds an MFA in visual arts (computer graphics) from Rutgers University.
Data visualization, design, and art in molecular biology
Dr Seán I. O’Donoghue, FRSC | CSIRO and Garvan Institute for Medical Research, Sydney, Australia
Molecular biology presents life scientists with two key challenges. The first challenge is extremely complex data – for this reason, many potentially important breakthroughs in biomedicine remain buried and undiscovered in publicly available datasets. Data visualization and graphic design – combined with classical bioinformatics approaches – can help bring these discoveries to light. Then, life scientists face the second key challenge: how to communicate these important yet complex, molecular-scale discoveries with peers, funding agencies, and the general public. Here again, principles of data visualization can help, together with animations that use approaches from cinematography, storytelling, and art. This talk will illustrate these issues using work from my team (http://odonoghuelab.org).
Seán O’Donoghue is an Office of the Chief Executive Science Leader in Australia’s Commonwealth Scientific and Industrial Research Organisation (CSIRO), Sydney. He is also Group Leader and Senior Faculty Member at the Garvan Institute of Medical Research in Sydney. He received his B.Sc. (Hons) and PhD in biophysics from the University of Sydney, Australia. Much of his career was spent in Germany, where he worked both in the Structural and Computational Biology programme at the European Molecular Biology Laboratory (EMBL), and at Lion Bioscience AG. He has been awarded a C. J. Martin Fellowship from the National Health & Medical Research Council of Australia, an Achievement Award from Lion Bioscience, and was recently elected a Fellow of the Royal Society of Chemistry. He is leading several projects, including: VizBi, an international initiative aimed at improving data visualization and user experience in bioinformatics software; VizbiPlus, aimed at educating and inspiring the general public about cutting-edge biomedical research (Finalist, 2015 NSW Emerging Creative Talent Award); Aquaria (first prize, 2015 NSW iAward for Research and Development), a resource that simplifies discovery and insight from protein structures; and Reflect, a browser plug-in designed to help in understanding life science literature that is widely used by many life scientists (first prize, http://www.elseviergrandchallenge.com).
From discrete event simulations of distributed database systems, to particle systems and then on to predictive analytics for social media, the math that ties all of these together
Alain Chesnais | TrendSpottr
This talk will cover the underlying mathematical concepts that we have used throughout our career to cover such varied topics as evaluating the probability of being in a deadlock situation in a massively distributed database using only locally available information, handling the probabilistic behaviour of large scale particle systems to model fluids and gases and then go more in depth into our current research into predictive analytics in social media to generate early warnings of what content is about to go massively viral.
Alain Chesnais is co-founder and Chief Scientist at TrendSpottr, which develops web services to identify real time trends in social media such as twitter and Facebook. He was the CTO of SceneCaster.com before then and previously served as director of engineering at Alias|Wavefront managing the team that received an Oscar for developing the Maya 3D software package.
Chesnais studied at l’Ecole Normale Supérieure de l’Enseignement Technique and l’Université de Paris VII and XI, and where he earned a Maîtrise de Mathématiques, a Maîtrise de Structure Mathématique de l’Informatique, and a Diplôme d’Etudes Approfondies in Computer Science.
He is a long time ACM volunteer and and served as President of the association from 2010 through 2012. Prior to his election as ACM president, Chesnais was vice president from 2008 through 2010 as well as secretary/treasurer from 2006 through 2008. He also served as president of ACM SIGGRAPH from 2002 through 2005 and as SIG Governing Board Chair from 2000 through 2002. He is a French citizen currently residing in Canada and has more than 30 years of management experience in the software industry.
Machine Learning with NVIDIA GPUs
Mark Harris | NVIDIA
NVIDIA GPUs are powering a revolution in machine learning. With the rise of deep learning algorithms, in particular deep convolutional neural networks, computers are learning to see, hear, and understand the world around us in ways never before possible. Image recognition and detection systems are getting close to and in some cases surpassing human-level performance, and deep convolutional neural networks have even been used to generate novel images and other content. I will talk about machine learning and the impact it is having in a variety of fields in the context of several NVIDIA initiatives ranging from hardware platforms to software tools and libraries.
Mark is Chief Technologist for GPU Computing Software at NVIDIA. Mark has fifteen years of experience developing software for GPUs, ranging from graphics and games, to physically-based simulation, to parallel algorithms and high-performance computing. Mark has been using GPUs for general-purpose computing since before they even supported floating point arithmetic. While a Ph.D. student at UNC he recognized this nascent trend and coined a name for it: GPGPU (General-Purpose computing on Graphics Processing Units), and started GPGPU.org to provide a forum for those working in the field. Mark lives off-grid with his wife and daughter in the MacKellar Range of northern New South Wales.
Professor Eleanor Gates-Stuart | Techno Art Program, National Cheng Kung University Taiwan
‘Mmm.’ relates to creative thinking in art and design practice, the intellectualising of thought and research involved in developing an idea and critical reflection in making work. In developing a new masters program in Techno Arts at the National Cheng Kung University (NCKU) Taiwan, the current intake of students have little general art practice experience but rather a group of individual skills that are shared in their research and desire for making things. Our aim is to find the core elements that trigger creative engagement and experimentation, transferring ideas into artistic output led by interdisciplinary research through the means of science and technology.
Mmm. signifies that hesitant stage in thinking towards action, the human mind processing meaning and rationalising an idea before being spoken. Such intense scrutiny, the surge of brain activity and reasoning is the simple foundation for generating communication amongst a diverse group of creative people, artists and scientists alike. This is not a discussion of philosophical values but a simple intrinsic method for establishing dialogue in an interdisciplinary team with a wide range of experiences and differing research fields. The simplicity of Mmm. creates an open space for dialogue, an interchange of ideas and entrepreneurial synergy for crossover practice.
Dr Eleanor Gates-‐Stuart, interdisciplinary artist and Professor of Techno Arts at NCKU, will discuss her research practice in the area of science and arts, her background in science communication, working with science organisations and her interest in promoting science innovation through her artist practice. ‘Art and Science as Creative Catalysts’ (2015) represents an insight into some of her collaborative research at CSIRO. She has been awarded Scitech’s Innovation in Art Residency 2016.
The secret life of small data in cities
A/Professor Martin Tomitsch | University of Sydney Australia
With big data being hailed as remedy to all the challenges of mass urbanisation, it is easy to forget that much of what makes today’s cities work is founded on the clever use of small data. This talk will use examples to uncover the history and evolution of small data in cities. It will specifically focus on the urban interfaces that shape the everyday experiences of people living in cities.
Associate Professor Martin Tomitsch is Head of the Design Lab and Director of the Bachelor of Design Computing at the University of Sydney, founding member of the Media Architecture Institute, and state co-chair of the Australian Computer-Human Interaction Special Interest Group. His research investigates the application of design principles and digital technologies for improving life in cities.
A New Approach to Create Integrating Art Therapy Using Animation
Adjunct Professor Jinny Hyejin CHOO | Korea National University of Arts / Chair at KoreaGraph
This research explores that animation can be used as an art form in an integrative approach of art therapy, and it includes a case study of drawn from the author’s own extensive work with animation in therapy using various media. Art Therapy helps individuals to externalize various psychological problems through creative art activities chosen to support a range of therapeutic methods. In recent years, art therapy has used drawing, music, writing, sound, movement and a variety of other art forms, acknowledging the intimate connection between numerous integrated theories of the arts and the promotion growth and healing. Animation, which focuses on drawn movement, is a unique art form; its production process incorporates many different art forms and media in a multimodal manner. In addition, using animation as a therapeutic approach could provide better access to children and adolescents, who generally understand and enjoy animation, and are familiar with the digital media environment. Ultimately, the integrated art therapy tools will extend the use of existing art therapy to digital media and pave the way for a more positive usage of new media. As this research is still at an early stage, many challenges remain. However, the case study demonstrates that animation therapy has great potential as a new approach to integrative arts therapy; offered in a supportive setting, it can intensify healing and transformation.
HyeJin(Jinny) CHOO received an MFA in Art and Film from the Graduate School of Advanced Imaging Science, Multimedia, and Film, Chung-Ang University, Seoul in Korea and a PhD from the same university in 2015. She majored in Illustration and Animation, and started her career as a freelance artist and producer in animation and media arts. She has been spending seven years in academia teaching animation and its related fields including Media Arts as an Adjunct Professor at the Korea National University of Arts in Seoul, Korea where she has also carried out research in integrated art education and art therapy using animation. She has used animation techniques in workshops and clinical practice since 2010.
Organizing a festival is of special interest to her alongside education and therapy in animation. She has chaired or participated in several international film festivals including SIGGRAPH Asia(2008, 2010), the GISF SF Festival 2011, the Seoul International Cartoon and Animation Festival(SICAF, 2003-2008) and the Puchon International Student Animation Festival(PISAF, 2000-2002, 2013-2014). She is currently involved in the Korean Indie-AniFest as a programmer, and has served as a computer animation festival director for SIGGRAPH Asia.
The art of visualisation at CSIRO
Dr John Taylor | CSIRO, Australia
I will provide a brief personal history of the impact of the art of visualisation on the process of scientific discovery. I will focus on the ability of visualisation to generate scientific understanding and to create engagement with a wide audience. As data volumes in science have rapidly increased, visualisation and visual analytics have now become essential tools for extracting knowledge from these massive data sets. Most scientific data is well structured arising from planned experiments and engineered systems. Unstructured data, for example, images, from scientific experiments and particularly of the real world, are now a major source of “big data” and are the subject of analysis in their own right using new approaches based on machine learning techniques such as deep neural networks. I will provide examples from recent CSIRO research projects that illustrate the importance of visualisation to gaining scientific insight and to conveying the results of complex scientific research to a wide audience.
Dr Taylor is currently Leads the CSIRO Computational and Simulation Sciences Future Science Platform and is CSIRO Data61 Research Group Leader, Computational Platforms. Dr Taylor has written more than 140 articles and books on computational and simulation science, climate change, global biogeochemical cycles, air quality and environmental policy, from the local to the global scale, spanning science, impacts and environmental policy. Dr Taylor’s research has been widely cited and attracted significant media attention. Dr Taylor has previously worked as a Computational Scientist and group leader both at the Mathematics and Computer Science Division, Argonne National Laboratory and at the Atmospheric Science Division at Lawrence Livermore National Laboratory. Dr Taylor was Senior Fellow in the Computation Institute at the University of Chicago. Dr Taylor has served on the Advisory Panel of the Scientific Computing Division of US National Center for Atmospheric Research (NCAR) and the US National Energy Research Scientific Computing Center NUGEX Advisory Committee. Dr Taylor currently serves on the Board of the National eResearch Collaboration Tools and Resources (NeCTAR) an Australian Government SuperScience initiative. Dr Taylor is a Fellow of the Clean Air Society of Australia and New Zealand.
Dr Caitilin de Bérigny | University of Sydney
I discuss artworks using mind computing technologies from the Digital Wall at Central Park. In this urban space the artworks invite users to record their brainwave data via a biosensor headset which captures their individual brainwaves, subsequently transforming the imagery displayed and creating an interrelationship between the wearer, the environment and the artwork.
Dr Caitilin de Bérigny is a Lecturer in the Design Lab. Her research in interaction design seeks to engage users in urban space. Caitilin is leading the Health & Creativity Node at the Charles Perkins Centre. Caitilin has been awarded numerous grants, including an ARC DP in 2016-2019 to design an exhibit for the American Museum of Natural History in NY, and National Museum in Canberra. She has exhibited and published widely. Her artworks have been exhibited internationally in 2015 in Germany at Center for Art and Media Karlsruhe; in France at the ENSBA in Paris and Marseille; in the USA at the Commencement Gallery; in Canada at McGill; in Australia at the National Gallery of Australia and in Berlin at the Wekstad. She won scholarships to undertake a Post-Diploma at the L’École des Beaux-Arts in Marseille, France and exchange program at the L’Écoles Nationale Supérieure des Beaux-Arts in Paris, France. She has studied and worked internationally in France and the USA.
The Three Views of “I See”
Professor Kerrie Mengersen | ACEMS & Queensland University of Technology
Immersive environments and virtual reality are providing exciting new ways of seeing our world. The challenge now is how to move from the “ooh” of IE and VR to the “ahh” of problem solving. This can be cast as three types of “I see”: creating IE and VR for a specific scientific and/or social problem, eliciting information seen by experts in these environments, and using this information to gain insight into the problem of interest. We consider this trifecta in the context of statistical modelling, whereby elicited information is formulated as a prior and merged with available data, and we ask the question: how much better do we do by bridging the gap between these new visual and statistical capabilities?
Professor Mengersen is a Professor of Statistics in the School of Mathematical Sciences and Institute for Future Environments at QUT. She has around 25 years of experience in statistical modelling, analysis and computation, with particular focus on applications in health, environment and industry. Her expertise includes analysis and integration of complex datasets, encapsulation and effective use of expert information, spatio-temporal analysis and complex systems modelling. In addition to academic outputs comprising over 250 journal articles, she has a continuous record of commercial consultancies with selected relevant clients including Corrs Chambers Westgarth (risk), Goronickel (design), Qld Environmental Protection Agency and Healthy Waterways (water quality), Port of Brisbane (prediction), Qld Dept Natural Resources (environmental statistical modelling and analysis), Western Mining Company (analysis) and Dairy Australia (triple bottom line sustainability). Professor Mengersen is a Deputy Director of the ARC Centre of Excellence in Mathematical and Statistical Frontiers for Big Data, Big Models and New Insights (ACEMS), and an ARC Laureate Fellow.
3D Capture and Annotation for Biological and Cultural Heritage
Matt Adcock | CSIRO Data61
Natural history collections are an invaluable resource housing a wealth of knowledge with a long tradition of contributing to a wide range of fields such as taxonomy, quarantine, conservation and climate change. However, such physical collections are often heavily underutilized as a result of the practical issues of accessibility. The digitization of these collections is a step towards removing these access issues, but other hurdles must be addressed before we truly unlock the potential of this knowledge. This talk will present some of the work CSIRO is doing to make 3D scans more useful and accessible to both scientists and the public.
Matt Adcock is a Senior Research Engineer, Experimental Scientist at CSIRO Data61. He has 15+ years of experience in Interactive Computer Graphics, Computational Imaging and User Experience Innovation (including 2 years at MIT Media Lab). He has contributed to or led a diverse range of projects, most recently working in collaboration with CSIRO’s National Biological Research Collections and the Atlas of Living Australia to develop a web-based platform for 3D annotation and sharing.
Capturing reality and creating a digital copy of the real world
Jon Baginski | Euclideon, Brisbane
Creating a digital copy of the real world and exploring in realtime provides interesting new possibilities for the visualisation industry.
We demonstrate some new technologies and processes for being able to convert and view extremely large datasets and utilise streaming technology to share instantly via the web.
Jon Baginski is currently Application Architect for Solidscan technology at Euclideon.
Having a cross discipline between technical art and data acquisition promotes unique methods for the digitisation of the real world. Jon began his career in visualisation by working in architecture followed by game development and interactive entertainment. After working in visual effects and discovering the process of LiDAR scanning and photogrammetry, Jon progressed to exploring real time visualisation of captured data, ultimately leading to the use and further development of the Euclideon Unlimited Detail Engine. The advent of next generation virtual reality hardware has also led Jon to explore new methods for virtual world building leading to unique and engaging experiences.