Interview Marty Altman (Hybrid Animation-Integrating 2D and 3D Assets)

Former cGI Technical Director, Walt Disney Feature Animation currently chief creative officer, Louisiana immersive technologies Enterprise

The convergence of all things “digital media," the evergrowing demand for content, and in particular shrinking budgets and time frames means the mixing of approaches and techniques is becoming more prevalent. Progressively more examples appear almost every day now, but this trend started years ago. Some specific examples of hybrid approaches from the not-too-distant past help tell this story.

The palace crowd element in Disney’s animated feature film Mulan was an interesting mix of 2D and 3D approaches. We utilized a slightly modified version of the proprietary crowd simulation, which was used for the wildebeest stampede on Lion King, the crowds on Hunchback of Notre Dame, the whales on Pines of Rome in Fantasia 2000, as well as the hun charge on Mulan. We also used a modified version of the editor written for the hun charge. The simulation placed the characters in 3D, and behaviors were simulated as before. The editor was a 3D application with certain 2D capabilities.

The 2D approach really came in with the palace crowd characters themselves. All the crowd behaviors, both cycles and transitions, were traditionally animated characters converted over to an interesting system of animated textures and mapped onto individual polygons. Colors for some portions of the textures, in this case skin tone and hair color, were locked. Other portions of the textures, such as clothing, used a false-color technique and art-directed color palettes to increase perceived variety.


This hybrid approach not only matched the traditional style of the film but provided a wide range of usefulness for the characters—they could appear full screen as could any other animated character in the film, and the same system could place 30,000 characters in a single scene. While still a good bit of work to traditionally animate the crowd characters, a pure 3D approach would have taken significantly more calendar time to meet those same requirements. If pure 3D had been the only option, we likely would have lost the palace crowd element due to budget and time constraints.

The Pooh’s Hunny Hut attraction at Tokyo Disneyland provides examples of integration issues going beyond just 2D and 3D, because of the practical aspects needed for a theme park attraction. One involves traditional painted backgrounds, a 2D character (Tigger) bouncing across three large screens, and a swarm of 3D characters (bees). It also has a big multiplane camera move incorporating both the projected artwork and several levels of practical elements placed between the viewers and the screens. For the multiplane shot to succeed, all these levels have to work together.

Another Hunny Hut example makes use of projection onto a half-silvered mirror, so viewers see themselves with a 3D character. Several issues made this challenging. A special projector was shipped straight from the factory to Tokyo, so we never had the chance to test our imagery directly and instead had to make adjustments based on phone conversations with the Imagineers in Tokyo looking at the screen. The heffalump is a semitransparent 3D character that fills up with semitransparent honey, and since the projector was fed with an NTSC source, we had to anticipate color issues. Of course, the hef-falump’s primary color was red (arguably NTSC’s least favorite color).

The short time frame and increased logistical issues required a different thought process. The heffalump element was broken down into several more rendered levels than would ordinarily be used, to provide maximum flexibility for controlling the final look in the composite. We also borrowed a bracketing technique from traditional photography to further cut down the need to go back and forth between Orlando and Tokyo to resolve the final version. A comment like “a third of the way between number 2 and number 3" meant one tweak of the parameters in the composite and we were done.

The convergence of animation, film, visual effects, broadcast, games, web and interactive, and other related industries is leading us to a place where the true power user not only prefers to have many tools in his toolbox but also has an insatiable need to learn more. The rationale is straightforward. The more tools and techniques you have at your disposal, the better your chances of seeing connections that lead to higher aesthetic quality, or to greater efficiencies, or both. Better quality gets you noticed and is always good. Efficiency, as in getting the job done on time and within budget, helps ensure you get the next call.

Next post:

Previous post: