Foundry at DigiPro 2019: Machine Learning in Post-Production Software


This post is by Jose Antunes from ProVideo Coalition


Click here to view on the original site: Original Post




DigiPro 2019

At DigiPro 2019, Foundry’s Head of Research Dan Ring will be presenting on “Jumping in at the Deep End: How to Experiment with Machine Learning in Post-Production Software”.

The 8th DigiPro, or Digital Production Symposium 2019, happens in Los Angeles on 27 July 2019, as part of SIGGRAPH 2019. The event is an open window to the future, as it  brings together the world’s premier creators of digital visual effects, animation, and interactive experiences. During the one day event, Scientists, engineers, artists, and producers share ideas, insights, and techniques that bring innovation to real-world production.

Foundry, the leading developer of

Filmtools Logo
DigiPro 2019
DigiPro 2019

software for the Digital Design, Media and Entertainment industries, reveals its plans for SIGGRAPH 2019, and those include participating at DigiPro 2019 Symposium. Foundry’s Head of Research Dan Ring will be presenting on “Jumping in at the Deep End: How to Experiment with Machine Learning in Post-Production Software”.

Behind-the-Scenes of Spider-Man

The DigiPro 2019 event will open with a keynote that sets the tone. Under the title “Swing Behind-the-Scenes of Spider-Man: Into the Spider-Verse”, Peter Ramsey, from Sony Pictures Animation and Danny Dimian, from Sony Pictures Imageworks, take the audience through the challenges of creating the big screen version of Spider-Man.

From animation to technology, artists were challenged with developing new tools and techniques to create a groundbreaking visual style for this fresh and highly original film, which introduces Brooklyn teen Miles Morales and the limitless possibilities of the Spider-Verse, where anyone can wear the mask. For this year’s keynote, co-director Peter Ramsey and visual effects supervisor Danny Dimian will discuss their experiences while making the film.

The program includes three sessions, The Nitty-Gritty, Learnings and Artist Tools, divided in a total of 10 segments.  Physically Based Lens Flare Rendering in “The Lego Movie 2”, Distributed Multi-Context Interactive Rendering, or Creating Interactive VR Narratives through Experimentation and Learning are some of the titles of the different presentations, a clear suggestion of the diversity of themes and techniques covered.

Machine-Learning in Post-Production

A team from Foundry, composed of Dan Ring, Johanna Barbier, Guillaume Gales and Ben Kent will participate in the session entitled “Jumping in at the Deep End: How to Experiment with Machine Learning in Post-Production Software”, which also includes Sebastian Lutz, from the Trinity College Dublin.

The information related to the session points to the fact that in recent years we’ve seen an explosion in Machine Learning (ML) research. “The challenge is now”, continues the note, “to transfer these new algorithms into the hands of artists and TD’s in visual effects and animation studios, so that they can start experimenting with ML within their existing pipelines.”

The session at DigiPro 2019  presents some of the current challenges to experimentation and deployment of ML frameworks in the post-production industry. It introduces Foundry’s open-source “ML-Server” client / server system as an answer to enabling rapid prototyping, experimentation and development of ML models in post-production software. Data, code and examples for the system can be found on the GitHub repository page.

DigiPro 2019

Foundry’s All Stars 2019

Foundry’s present at SIGGRAPH 2019 extends beyond the Digital Production Symposium (DigiPro 2019). Returning for the third year, Foundry’s own All Stars event will take place on Sunday, July 28th, with speakers from the most innovative companies from around the world including Digital Domain, DNEG, Nike, More VFX, LAIKA, Method Studios and Weta Digital. The event features an inspirational keynote from Marvel Studios and opening remarks from Mikki Rose, SIGGRAPH 2019 conference chair.

Foundry is also supporting the SIGGRAPH Student Volunteer Program, the annual JPR SIGGRAPH Luncheon and the new Pipeline Conference, taking place on Sunday, July 28th. Foundry’s Exhibitor Sessions will take place on Monday, July 29th and Tuesday, July 30th. They will focus on Education, Look Development and Lighting, Cloud-Based Solutions, solving creative challenges with Nuke and Modo in 3D design.

DigiPro 2019

Foundry at SIGGRAPH 2019

Jody Madden, Chief Product and Customer Officer at Foundry, commented: “SIGGRAPH is a great opportunity for us to meet our customers and share the latest updates on our products and innovations with the wider industry. We are thrilled to return this year with our third annual All Stars event with another impressive line-up of speakers sharing their work that truly brings imagination to life with our products. Complemented by a suite of exhibitor sessions on everything from education to research, we’ll have something for everyone.”

Mikki Rose, SIGGRAPH 2019 Conference Chair, commented: “I am very excited to be speaking at Foundry’s All Stars event this year, where I plan to address how our industry can continue to thrive and foster creativity by creating a more diverse and inclusive future for artists. Foundry’s All Stars speakers this year truly show how powerful software can inspire creativity in artists and studios from all around the world.”

Foundry will be located at Booth 925 during SIGGRAPH 2019, where they’ll be showcasing products in partnership with Lenovo and NVIDIA. Foundry’s team will also be present at the NVIDIA Limelight event on Monday, July 28th with the latest Modo updates and in the NVIDIA Booth on Wednesday, August 1st, discussing Machine Learning. Foundry’s products will also be showcased at the following booths: WACOM, NVIDIA and AWS.

The post Foundry at DigiPro 2019: Machine Learning in Post-Production Software appeared first on ProVideo Coalition.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.