From the glittering ground slowly emerges a mass of moving particles. Several others  – all alike – soon join the form, dancing as though luminous bodies in the infinite space of the cosmos. These precarious shapes constantly shatter, appear and disappear to the rhythms of the final transition of Stravinsky’s Rite of Spring…

They now move in an ocean of light, sinking in the glow and reborn as reddish, then bluish creatures; they oscillate from one color to another while the matter transforms and slips away in columns of sparkling fire, flowing and bubbling lava, bursts of coloured glass, flying particles, and aerial smoke…

Designed in 3D and also available in 2D, CODA draws on advanced digital technologies to offer a new vision of dance in cinema. Designers Denis Poulin and Martine Époque create virtual dancers free of their morphological appearance. They imagine a dance in which humans become dynamic footprints, carrying the motor signature of each one of the real dancers at the source of their movements.

With motion capture (MoCap) and particle processing, Denis Poulin and Martine Époque evoke nature and address environmental issues through metaphor by providing in a single gesture a contemporary reinterpretation of the Rite of Spring. From a bright cloud of particles a world is born – a universe that is also a digital stage wherein lurks the first dancer, whirling in the first environment… CODA is thus genesis and apocalypse, a demiurgic myth of the creation of the world, a fused universe where space and time collide, deploy, and dissolve.

Because motion is the core of a dancer’s identity, Denis Poulin and Martine Époque wanted to capture movements beyond the limits of the skin. To do so, the designers shot CODA using a motion capture (MoCap) system, from which resulted polygonal characters that accurately represented the dancers’ movements. However, rather than making computer-rendered characters covered with a skin texture, Denis Poulin and Martine Époque associated them with particles. The particles interacted with each movement based on the physical characteristics attributed to them – characteristics that differed in each chapter of the film. The particles thus enter and leave the body, fly in the wind or slip on surfaces like a liquid. This way, CODA offers a reinterpretation of movements, focusing on a more universal representation of the body in motion.

Designed in 3D and also available in 2D, CODA draws on advanced digital technologies to offer a new vision of dance in cinema.

On the origins of the project

Denis Poulin: I started in dance in 1968, and at the time I was also doing photography—and I quickly grew interested in filmmaking. I did a master’s degree in communications at the University of Michigan, and while I was there I directed three dance films, which I sent to Norman McLaren. He liked them enough to pass them on to producer Marc Beaudet, who immediately offered me a contract. That’s how I came to direct Beyond Curtains, my first film at the NFB. It was released in 1978, and was based on choreography by Martine.

I wanted to pursue this path, but it was very difficult to find funding for dance films. I continued to dance myself, and also taught dance and used video projection during dance shows. Martine and I even started a video dance program at Université du Québec à Montréal (UQAM). Around 1990, I started doing dance animation using 3D software, but it was not really sophisticated enough. Nonetheless, I still managed to make a small film using that technology. Things really started to move in 2000, with the arrival of motion capture (MoCap). We had been following the evolution of the technology, and when MoCap arrived it offered tremendous possibilities.

Martine Époque: I should also say that there was an important turning point for us in 1988, when I decided to quit the professional dance scene after becoming involved with the UQAM dance program. It gave me the means to explore dance and to perform research. I left behind organizations that funded dance in favour of those offering research grants. That’s how we wound up going from video dance to technochoreography, to MoCap—and how we came to develop a film based on the idea of dance without bodies.

On the technique

D.P.: When you do MoCap, you go from a person represented through polygons to one covered by an envelope. This envelope is the virtual person’s body. It could be anything from a small penguin to Captain Haddock… though obviously that’s not the direction in which we wanted to go. Our idea was to visualize dance in the most abstract way possible. That’s how we came to imagine people composed of particles of light.

Early on, I did some tests. The software tutorial led me to fill a glass with particles, and depending on their characteristics, they would react differently to falling out of the glass. I thought that what was true of the glass and particles would also be valid if I took the polygonal body as the vessel and did not create a skin for it. This would mean that the body would remain invisible—a transparent vessel—and would only take shape through its interaction with the particles.

Having established that, we had to find a variety of ways to emit and constrain the particles. That’s when we met Vincent Fortin, a real particle artist who had the necessary technical expertise. He was an essential part of bringing our research to life.

M.É.: We first had this idea of dance without bodies in 2004. Thanks to a grant from SSHRC (Social Sciences and Humanities Research Council), we created a prototype in 2008. It allowed us to recognize that what was most interesting about the material we were capturing wasn’t its corporeal aspect—lines, mass, bodies—but the energy. So from 2008 to 2010 we compiled what we called a Digital Collection of Signature Movements from different dancers. We used MoCap to record 13 well-known Quebec dancers, including Louise Lecavalier and Marc Boivin, and created a game which we made available on our website. We thought of then cloning them using avatars, but after thinking it over we decided not to go that route.

D.P.: A dancer’s identity is found in movement. The dancer’s body is the vehicle, but it is not their true signature. When we tried to clone, it felt as though we were usurping their identity. By eliminating the body completely, we approach a state of dance that I would call pure. It’s like looking at something through a microscope—we see the object differently. That’s what we wanted to do with the tools we had available: lead the viewer to see dance differently.

On The Rite of Spring

M.É.: It’s a long-standing love affair: it was one of the last pieces I choreographed professionally, and Denis was one of the three soloists. We went all the way to the cultural section of the Calgary Olympics with that show, and were even invited to go to China—which was interesting, since my Rite tells the story of a dictator overthrown by the people.

Later, one of my doctoral students told me that two twins were doing a PhD in music on The Rite of Spring. That’s how we came to meet Hourshid and Mehrshid Afrakhteh, who play the music in CODA.

At first, we’d had the crazy idea to do the entire Rite of Spring.

D.P.: We revised our thinking after talking it over with people from the NFB. They helped us understand that a film with a 33-minute running time was not suitable for broadcast, and that we would either have to make a film that was shorter or one that was longer. Doing a kind of “making of” to extend the length of the film didn’t really excite me, especially since it would make an already heavy process even heavier. So we decided to scale the project back to 10 minutes, concentrating on the finale. In musical terms, a coda is a reprise of certain phrases that have already been heard. It can be a sort of summary of a piece. It was in that spirit that we approached the last nine minutes of The Rite of Spring—as a summary of the principal tableaux.