DETAILS


COLUMNS


CONTRIBUTIONS


ARCHIVE



ABOUT THE COVER


Vol.33 No.4 Nov. 1999
ACM SIGGRAPH


Computer Vision Applications Astound the Imagination

Karen Sullivan
Computer Graphics Cover Editor

November Columns
From the Editor Gaming & Graphics

About the Cover
Previous About the Cover Previous About the Cover


Front Cover

Those of you lucky enough to catch Paul Debevec’s short film, Fiat Lux, in the Electronic Theater at SIGGRAPH 99 have already experienced a dynamic introduction to just one of the possibilities of fusing computer graphics and computer vision: capturing the world from photographs and video for realistic image synthesis in motion picture and scientific visualization. This issue of Computer Graphics aims to expand our knowledge base in the field by detailing the amazing advances and possibilities that are changing the way we produce computer imagery.

Happy happy Budda may soon be arriving on your desktop (the real one...) via your personal 3D fax machine. Brian Curless’ research on range scans to produce 3D models is illustrated on this issue’s front cover. The large background images are examples of active range sensing via a common form of optical triangulation. This is a scanning method for the creation of 3D data. His article within this issue, "From Range Scans to 3D Models," provides more detailed information. The bottom series of images shows recent results from Curless’ project to build a 3D fax machine. The images from left to right are:

  1. Digitized video images of original Budda statuette, rosewood and ivory.
  2. Statuette painted gray to simplify scanning.
  3. First scan: a Gouraud-shaded rendering of one range image of the statuette.
  4. 48 scans: Gouraud-shaded rendering of a polygon mesh created by integrating 48 aligned range images of the statuette. As each range image was acquired, it was converted to a volumetric representation and integrated into a volume representing the entire model.
  5. Final model: RenderMan rendering of the final model after application of an automatic hole-filling algorithm.
  6. Faxed hard copy.

A paper, "Volumetric Method for Building Complex Models from Range Images" was given at SIGGRAPH 97. More information on Curless’ research can be found at: http://graphics.stanford.edu/projects/faxing/happy/.

Back Cover

The back cover shows a sampling of images from our focus articles in this issue.

The top series of images are from an animated short, Cooking with Kurt. This series takes us from live video to reconstructed deformable vegetable models to animation frames for a physics-based vegetable sequence. More information on visual modeling can be found in Demetri Terzopoulos’ article, "Visual Modeling for Computer Animation: Graphics With A Vision," in this issue.

Eric Mortensen demonstrates new selection and compositing techniques, making the lion seamlessly lay down with the lamb in the left column of images. His article is "Vision-Assisted Image Editing."

The right column of images begins with the Morphing Mona Lisa contributions from our focus editor, Steven Seitz, and Charles Dyer.

Below this is one of 15 photographic references of the Berkeley campus used to construct a 3D Berkeley campus model for Paul Debevec’s Campanile movie. The film model also includes photogrammetrically recovered terrain geometry that extends out to the horizon. Please see his article, "Image-Based Modeling and Lighting," in this issue.

The middle bottom image is provided by W.T. Freeman, lead author of "Computer Vision for Computer Interaction" in this issue. It illustrates the use of a CMOS detector chip made by Mitsubishi Electric, called the artificial retina chip which can both detect the image and perform some image processing operations. This image shows The Nintendo Game Boy camera using the artificial retina chip to allow the user’s face to be inserted.

The bottom right image show 2D tracking: a process of following a feature as it moves through the images in a shot. In Interview with a Vampire, Brad Pitt’s makeup was too heavily applied in one scene. This wasn’t noticed until after the production had wrapped, making it impossible to refilm the scene. By 2D tracking the features on his face and compositing digital makeup, the effects of the actual makeup were reduced. For more details, refer to Doug Roble’s "Vision in Film and Special Effects" in this issue.


Karen Sullivan
Ringling School of Art and Design
2700 North Tamiami Trail
Sarasota, FL 34234

Tel: +1-941-351-5100


The copyright of articles and images printed remains with the author unless otherwise indicated.