What’s happening at Siggraph Asia next week

Next week is the start of SIGGRAPH Asia. Featured sessions are the curated content of inspirational work and speakers from around the world. The program casts a spotlight on major breakthroughs, techniques, and arts in the field of Computer Graphics and Interactive Techniques.

There’s still time to register for the conference and you can get 10% off with our exclusive fxguide discount code. Just enter the code at checkout: FXGUIDE10GK0RC

Visit the SIGGRAPH Asia web site to register now.

Joe Letteri (Wētā FX)

This is an exciting year for SIGGRAPH Asia, the world has opened up after COVID and technology has opened up with advances in AI and Machine Learning, this year’s speakers cover the full range of Animation and visualisation disciplines. It is the arguable one of the most impressive lineups of featured speakers SIGGRAPH Asia has ever had. Multi-Oscar winner Joe Letteri, Senior Visual Effects Supervisor, Wētā FX is providing a Keynote presentation on day two of the conference which will no doubt be one of the most anticpated talks. Joe Letteri recently spoke at DigiPro/SIGGRAPH LA and his depth of knowledge and devoition to impactful storytelling is legendary. The other featured speakers include some of the world’s leading experts in R&D and production from around the world, including ILM, Wētā FX, Netflix, NVIDIA, Meta and Soul Machines.

The Intersection of Art and Technology: Creative Design at ILM.

Rob Bredow, SVP, Chief Creative Officer at Industrial Light & Magic.

Rob Bredow is very well known not only as an Academy Award-nominated VFX Supervisor but also as Chief Creative Officer of IILM. Bredow focuses on the creative strategy for the storied visual effects company, where he is responsible for the company’s overall creative teams, innovation, and business development. Rob also serves in the role of SVP; Creative Innovation for Lucasfilm. Some of Bredow’s previous projects include Solo: A Star Wars Story, ILMxLAB’s story-based virtual reality experiment Trials on Tatooine, and films dating back to Independence Day.

Solo: A Star Wars Story

Graphics and AI: Better Together

Bryan Catanzaro, Vice President of Applied Deep Learning Research, NVIDIA

Neural rendering is arguably the present and future of graphics. Today’s most advanced games use DLSS neural rendering to infer seven out of eight pixels and make real-time path tracing possible, significantly improving image quality, fluidity, and responsiveness, and Catanzaro helped create DLSS at NVIDIA.  Conversely, graphics are essential to the future of AI, because virtual worlds will be where people interact with AI for work and play. In this talk, Bryan will lay out efforts that NVIDIA is deploying to push graphics and AI forward jointly.

Bryan Catanzaro (R)

Dr. Bryan Catanzaro is Vice President of Applied Deep Learning Research at NVIDIA, where he leads a team of AI researchers working on multimodal language modeling, chip design, audio and speech, graphics, and vision, with the goal of finding practical new ways to use AI for NVIDIA’s products and workflows. In addition to DLSS,  he headed the Megatron project, which set speed records at scale for training large language models and forms the technical basis for many LLM projects around the industry as well as NVIDIA’s own Generative AI efforts.

Blood, Guts and Fur: The Visual Effects Behind Cocaine Bear

Robin Hollander, VFX Supervisor,

VFX Supervisor Robin Hollander will share his insight into the outrageous story of Cokey, a bear who accidentally ingests a ton of cocaine and will do anything to get her paws on more.

Developing Cokey’s character required a significant amount of animal exploration, including early motion studies to ensure the portrayal of our coked-up bear felt as realistic as possible. Robin and his team took learnings from other animals by conducting a series of behavior studies that were fundamental to giving our bear her unique animal quirks.

Throughout Cokey’s chaotic rampage to find more drugs, the audience sees four increasingly disheveled and blood-thirsty ‘bear looks’. During the talk, Robin will share details of how his team designed, built, and animated the giant black bear – from her body and fur to her teeth, claws, and various bloody grooms. One of the critical parts of animating Cokey was capturing her interactions with the on-set actors. Robin will touch on how the team partnered with an on-set performer for interactions and developed countless groom refinements to help perfect Cokey’s overall look and feel in the action-packed, over-the-top thriller.

The Next Tools.

Mark Pesce, Honorary Associate, Media and Communications, University of Sydney, who recently authored the book Augmented Reality: Unboxing Tech’s Next Big Thing by Cambridge Press.

Despite foundations deep within computing, many tools used in VFX remain manual, difficult, and obscure. It can often take years of practice to become expert in modeling or animation, resulting in a field full of deep experts, but one surrounded by a moat that limits some from creatively contributing. Two recent works from Australian researchers point toward the advent of a new generation of tools that harness artificial intelligence in a balanced tension between expertise and creativity. 3D-GPT eschews “the ‘diffuser’ models for a deeper and more complex tool chain, providing a rich, linguistic interface to a parametric tool” he explains. CICADA shows how drawing can be redesigned to “provide ‘space’ for both humans and AI to co-occupy and co-create in a manner that enhances collaboration. Combined and generalized, these efforts define a template for a new class of creative tools that ‘handle the boring bits’ while simultaneously acting as handmaidens and amplifiers of human creativity”.  Pesce in his talk will show how these ‘Next Tools’ can bridge the moat between creativity and effective computer-supported communication.

From Virtual Cinematography to Virtual Production

Dr. Paul Debevec, Chief Research Officer, Eyeline Studios Powered by Netflix

Paul Debevec is well known to both the SIGGRAPH community and readers of fxguide. This talk will describe how virtual cinematography techniques developed at UC Berkeley for image-based modeling, rendering, and lighting helped enable iconic visual effects sequences in movies such as The Matrix, X-Men, Spider-Man 2, Benjamin Button, and Avatar. It will also show how real-world image-based lighting techniques which surround actors with computer-controlled LED’s have led to new virtual production techniques seen in The Social Network, Gravity, Rogue One, Asura, and The Mandalorian. The talk will conclude by describing new research to improve the lighting reproduction, color rendition, and alpha compositing capabilities of these new virtual production stages.

Unraveling The VFX Behind The Marvels

Sean Walker, VFX Supervisor and  Sabine Laimer, Sequence Supervisor: Wētā FX

A quest to save the universe and an entanglement of powers form the backdrop for Carol Danvers’ enthralling journey in the MCU’s latest, The Marvels. In this featured session, Wētā FX’s VFX Supervisor, Sean Walker, and Sequence VFX Supervisor, Sabine Laimer, will share their experience working with Marvel to bring the third act to life.

The Marvels

Leading a cross-departmental team of artists, Sabine will explore how initial post-production began with the team building sequences on their mocap stage, combining the action beats with new concept art, and pre and postvis to give new energy to the story. Working on nearly every MCU film Wētā has been involved with, Sean brings a history of Marvel knowledge and is an expert in bringing together photoreal VFX tenpole films. With production taking place during the pandemic, Sean will discuss how the use of high-quality digi-doubles was integral to keeping this film on track and fulfilling the vision for the film. As their power entanglement creates for a chaotic third-act fight-sequence, FX work was key to identifying each of our hero characters amongst the fast-paced action. Sabine and Sean will touch on how they developed the unique look for each powerset as they battle to save the world.

Panels

We have panels on both Character Animation and Digital Humans exploring the state of the Art and looking forward to what is being developed.

Digital Humans Panel: How We View Digital Humans

Christophe Hery, Director, Science Research, META , Mark Sagar, CEO and co-founder of Soul Machines and moderated by Fxguide’s Mike Seymour

Digital humans are a multidimensional problem with vectors in interactivity, personalization, scalability, realism, and engagement. This panel will explore

  • What are the main issues facing creating useful digital humans
  • Exploring realism, behavioral simulation, and ethical considerations?
  • The impact of generative AI on digital humans
  • The fork in the reproduction of digital humans -between traditional CG and A.I.

The panel discussion will focus on how the look of digital humans is contextual and what it takes to make a Digital Human authentic.

Animation Panel: The Craft of Character Story Telling in an Age of High-tech Tools 

Raqi Syed, Senior Lecturer, Victoria University of Wellington, Emily Dean, Director, Blur Studios & Nexus Studios, Rob Coleman, Creative Director, Industrial Light & Magic and moderated by Fxguide’s Mike Seymour.

Who is the ‘actor’ making the acting choices in an animated film, is it the voice actor? the animator? the director? Performance is both technical and artistic and this panel will explore how these work to produce a powerful and effective performance.

This panel aims to look at the ways animators effectively explore and deliver subtext in character animation. The panel will discuss how generative AI may affect performances moving forward and the developments that the panel sees being required in coming years