GDC + [email protected] 2018 Session Scheduler

View, browse and sort the ever-growing list of GDC sessions by time, pass type, track, and format. With GDC Session Scheduler, you can build your schedule in advance and access it during the show via export or with the GDC Mobile App. Sessions do fill up, so please arrive early to sessions that you would like to attend. Adding a session to your schedule does not guarantee you a seat.

In order to best utilize Session Scheduler, please use the login credentials you've set up on this page as your login credentials on our GDC Mobile App. Your login credentials are the same as your registration login. You can download the app on iTunes or Google Play.

Production-Level Facial Performance Capture Using Deep Convolutional Neural Networks


Antti Herva (Lead Character Technical Artist, Remedy Entertainment)

Location: Room 3005, West Hall

Date: Wednesday, March 21

Time: 3:30pm - 4:30pm

Pass Type: All Access, GDC Conference + Summits, GDC Conference - Get your pass now!

Topic: Programming

Format: Session

Tutorials: N/A

Vault Recording: Video

Audience Level: Advanced

In this presentation Antti Herva, Lead Character Technical Artist at Remedy Entertainment will present a machine learning solution that enables cost efficient creation of large amounts of high quality facial animation for digital doubles in games. To this end, Remedy Entertainment, Nvidia and the University of Southern California's recently researched "Production-Level Facial Performance Capture Using Deep Convolutional Neural Networks" was published as part of the Symposium on Computer Animation in 2017. Topics covered in this session include recording a facial animation dataset for an actor, setting up a deep learning project, preprocessing the data, training a deep convolutional neural network, evaluating the results, a summary of the findings and a discussion on potential future work.


Gain insight into machine learning for content creation. In this talk a fairly conventional and popular deep neural network architecture is applied to generate facial motion. The presentation is structured so as to help the audience understand the process of setting up a supervised learning project and the benefits and limitations of using machine learning to solve production issues in AAA games.

Intended Audience

A variety of people involved in the creation of convincing character animation will find this this session interesting, including but not limited to programmers, producers, researchers, technical artists and animators. Anyone interested in machine learning for content creation where high quality and fast turnaround are required will likely benefit as well.