Capture Captation

Computer Science studies
Rémi Ronfard, Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LJK, 38000 Grenoble, France

I have an educational background in engineering and I’ve spent my entire career in computer science research, first in the industry then with Inria [National Institute of Research in Science and Technologies] where I currently work as a research director. Among my numerous centers of interest, “capturing” theater has always aroused my curiosity, even if I don’t particularly like the word, which evokes an act of diversion or theft. I prefer to consider it as an instrument of memory and sharing. I have been particularly struck by a question asked to René Allio and Antoine Bourseiller by Jean-Luc Godard in an old edition of Cahiers du Cinéma: “Why don’t theater people ever want to film their shows in order to preserve them as archives? One extra day of acting, and the takings of day would serve to film the show. It would be very simple: the camera in the middle of the orchestra, with a medium lens – but no zoom, which would give way to interpretation.”

Why, in fact? Why are televised productions of theater met with so little success? Why are they often so poorly filmed? Is there something within theater that irreducibly resists cinematographic capture? Or is there a way of filming, that awaits invention, that could capture and restore or reproduce theater in a more satisfying manner?

I began to imagine devices that would permit us to go beyond the limits of classical points of view. My first idea was to capture the theater in three dimensions. Consequently, we set up at Inria a 3D reconstruction system that was quite elaborate and that was put to use in the laboratory with the help of many cameras. Would it be possible to launch this same system around a theater stage in order to integrally reconstruct the performance? After a few preliminary attempts, this first project was rapidly abandoned due to its astronomical cost, which could not be sufficiently justified by the advantage of seeing the stage from multiple points of view. After all, theater isn’t watched from the stage wings, nor do we change our seat with each scene.

In a second attempt, I adopted a diametrically opposed approach that consisted in filming the play from a single, privileged point of view (the ‘Prince’s Seat’) with the highest resolution possible, in order to then be able to digitally zoom-into the image and create new, previously unseen spatial and temporal compositions, ‘split-screens’ and multi-screen projections (Ghandi, 2014, Kumar, 2017). This research program was enthusiastically supported by the Théâtre des Célestins in Lyon, which allowed me to act as thesis director for Vineet Gandhi, now professor at the International Institute of Information Technology in Hyderabad.

This project was underway when Julie Valero asked me to film rehearsals for the performance “The Making of Monsters” by Jean-François Peyret in Grenoble and Lausanne. With these filmic captures, it would be possible to later create audiovisual notebooks, associating text and image in an unprecedented way (Ronfard and Valero, 2020). Within the scope of the Performance Lab, we have made all of the necessary technologies for the creation of these audiovisual notebooks available in the form of an online service (kinoai.inria.fr). And our research projects now continue in new directions, such as stereoscopy and automatic editing. Theatrical capture has as such become a privileged thematic for my research team.

Theatrical capture should be distinguished from motion capture. The rapid evolution of artificial intelligence, however, suggests that it will soon be possible to reconstruct actors’ movements in 3D from a video recording. And we will have come full circle.

More from this author:

Vineet Gandhi, Rémi Ronfard, Michael Gleicher, “Multi-Clip Video Editing from a Single Viewpoint”, European Conference on Visual Media Production, 2014 [online] : https://hal.univ-grenoble-alpes.fr/hal-01067093v1 (03/11/21)

Moneish Kumar, Vineet Gandhi, Rémi Ronfard, Michael Gleicher, “Zooming On All Actors: Automatic Focus+Context Split Screen Video Generation”, Proceedings of Eurographics, Computer Graphics Forum, Wiley, 36 (2), 2017, pp.455-465 [online] : https://team.inria.fr/imagine/zooming-on-all-actors-automatic-focuscontext-split-screen-video-generation/ (03/11/21)

Rémi Ronfard, Julie Valero, “KinoAI et le carnet audiovisuel : une solution plurielle pour l’étude des répétitions”, European Journal of Theatre and Performance, European Association for the Study of Theatre and Performance, 2020, pp.334-375 [online] : https://hal.inria.fr/hal-02797350/document (03/11/21)

Cite this item: Remi Ronfard, “Capture”, translated by Lauren Fabrizio, Performascope: Interdisciplinary Lexicon of Performance and Research-Creation, Grenoble: Université Grenoble Alpes, 2021, [online]: http://performascope.univ-grenoble-alpes.fr/en/detail/177969