Eyeing Disney Zurich Research
Digital Emily 2.0 and the great discussion that surrounded the publishing of her data and renders lead to the decision to generate a second data set, this time with a member of the DHL as the subject. This would allow us as a group to develop and run a whole new range of scans and facial captures. I was suggested as the subject for the new scans.
The first of these new scans was done at USC-ICT. This was incredible successful and the results from that session are still being processed. I was scanned at USC-ICT in August in one of the most complete scan’s that team have ever done. The reason I was chosen was simple, I was readily available for any scanning work pretty much anytime, and I am well enough known that the work we do as a group can be judged not just on ‘life-likeness’ but also on its accuracy to me, – after all the years of fxguide and Wired magazine videos, we guessed my face is now well known in the effects community. (We will post more about the first, separate USC-ICT scanning in an upcoming blog)
For this new second stage we have just been lucky enough to have been invited to have new face scanning done at Disney Zurich Research. The aim was very different from the USC-ICT data set. Firstly, Disney Zurich Research has one of the most advanced eye scanning research in the world, and apart from some reference imagery, we did not have a true scan of my actual eyes. We also got to capture static scans and animated dialogue using the Disney Zurich Research team’s Medusa rig. Below are images taken during the session in Zurich earlier this month.
The team in Switzerland were both generous and great to work with. The Eye work was headed up by Pascal Bérard, who is doing his PhD with Disney Research Zurich focusing on eyes. I also worked closely with Derek Bradley, Research Scientist and Thabo Beeler, Research Scientist, who leads the Head Capture & Effects.
Disney Research Zurich is so well known to me from their published research work, but I had never visited the team before and so I was naturally very keen to take up this chance to work with one of the most respected graphics research teams in the world.
A fish eye view of what one sees when lying in the eye rig, note the primary main 6 cameras which all focus on one eye at a time.
You can clearly see my eye in the back of each of the DSLRs during this alignment stage
During the actual scan different LED lights are used as part of the process
To remain still for the session it has been proven helpful to have a neck and head brace.
The team also calibrate the images after each session
(…and if you are wondering how bloodshot my eyes are after 20 hours flying)
Here are links to the published research that is the basis of what we were doing in Zurich with eyes.
Facial movement and scanning: Medusa Rig
The Medusa Performance Capture system, developed by Disney Research in Zurich, consists of a mobile rig of cameras and lights coupled with proprietary software that can reconstruct actor’s faces in full motion, without using traditional motion-capture dots. The technology comes as the result of many years worth of research and scientific advances in capturing and modeling of human faces.
Medusa delivers high resolution 3D faces, with the ability to track individual pores and wrinkles over time, providing very realistic facial geometry that is ideal for creating digital doubles for visual effects and computer games.
Above I am being scanned by Disney Research’s impressive Medusa Rig
The Rig not only is used for the shape of the face but for frame accurate animation.
This rig is both portable and lightweight. Newer versions are also planned.
We recorded both FACS poses and dialogue sequences as one would do in a typical Medusa Session.
Industry professionals at major VFX facilities have already included Medusa data in several major Disney Hollywood feature film face pipelines.
Here is the link to the Medusa research: http://medusa.disneyresearch.com/
After the Medusa I had a static head scan from a portable Disney Research Rig, which works on Photogrammetry
This mobile rig has been extensively used by the research team around the world.
We have provided links to the published research that the team has done in each of these areas above, and we look forward to sharing more with you in due course. Again I’d like to point out how very generous it was of Disney to allow us such extended access and I really want to thank the entire Disney Zurich Team for their warm welcome and effort in doing this work for the Wiki Human Project and the DHL.