Sony Pictures Imageworks rose to the dual challenge of creating a swinging Spider-Man travelling through a virtual New York. All images unless otherwise noted are © 2002 Columbia Pictures Industries, Inc. All rights reserved.

"Spider-Man -- Spider-Man -- doing whatever a spider can."

In regards to the approach the effects team took on the new live-action blockbuster Spider-Man, the old theme song surely rings true. The folks at Sony Pictures Imageworks were given the daunting task of making a man swing through gulches of city buildings, crawl up vertical walls, dodge super-charged throwing stars, along with various other feats of wonder, in nearly 500 effects shots. To add to the mix, it had to look as real as possible. Superhero flicks of yesteryear have been set in fictional surroundings -- Spider-Man was set in the very familiar movie location of New York City. I talked to the film's animation director Anthony LaMolinara and visual effects supervisor Scott Stokdyk to find out the process in which the effects team approached bringing the man in red and blue to the big screen.

Spider-Man's animation director Anthony LaMolinara (left) and visual effects supervisor Scott Stokdyk were key to the film's animated visual effects. © Sony Pictures Imageworks.

Re-Creating Reality
One of the toughest challenges was the creation of the virtual New York City setting. To achieve the appearance of Spider-Man swinging from building to building, moving away from the camera and toward it, the filmmakers were forced to create the landscape in the computer due to the impossibility of shooting the sweeping camera movements in real life. To get the look of the city as perfect as possible, the effects and production design team started off by surveying both Times Square and the Queensboro Bridge to gain accurate geometry and texture details. LaMolinara described part of the process of creating the cityscape as a pan and tile procedure. Approximately 25 images (or tiles) of the background environment were taken along a 180-degree axis and then were stitched together to create a half sphere for the action to move through. Afterward, other buildings were placed in the foreground and rendered as complete 3D objects. If you're asking yourself why this is interesting and why it's significant I'll tell you.

To create a complete city of 3D buildings would have taken an enormous amount of computer space and time. This tile and pan process saved time and memory space by limiting the amount of complete 3D items. The reason the flat background works is because of the way one's eye registers perspective. When the camera moves past objects the perspective changes greatly when things are closer. However, the perspective of objects in the far background would stay relatively the same -- thus the flat image is unnoticeable when moving through the foreground space.