Depth perception: an entertainment store uses 3D music videos and film trailers to attract customers. (digital video).

To draw the attention of holiday shoppers, retailer Wherehouse Music used Dynamic Digital Depth's (DDD) 3D television system to showcase a selection of music videos and movie trailers in three-dimensional format.

"Watching content in 3D is still a unique experience for the average consumer, so a person tends to watch promotional material much longer if it's in 3D because the look is so different," says Renee Geddis, vice president of marketing and advertising for Wherehouse Music. "Our hope is that if we can attract attention to these clips, the consumer will be more likely to purchase the music or DVD."

Wherehouse Music provided DDD (Santa Monica, CA) with several music videos from artists such as Madonna, Britney Spears, and Alicia Keys, along with film trailers from Dr. Seuss' How The Grinch Stole Christmas, Shrek, and Jurassic Park 3. DDD then used its TriDef technology to convert the 2D video content into imagery that could then be viewed in 3D without the need for special glasses to see the stereoscopic effects.

"Everything you see on the screen [in 3D] started out in 2D," says Chris Yewdall, DDD's president and CEO. "We use a sophisticated set of tools in the postproduction process that allows us to segment material filmed with one camera, and recover depth information to create both left- and right-eye views for depth."

Presenting the 2D images in 3D required a new content-creation process and a delivery mechanism. To create the stereo appearance, graphic artists at DDD digitized the original color material and converted it to grayscale, creating a depth map. To generate the depth maps they first selected the first and last "key" frame of each video, and assigned depth to objects in those frames based on their distance from the camera--those closer to the camera were lighter shades of gray, while those farther away were darker. To accomplish this, the artists used DDD's proprietary DepthMapper software to manually "airbrush" several spots of depth onto the images in those two frames, rather than precisely outlining each object. DepthMapper--which can distinguish among 256 shades of gray, far more than the human eye can perceive--searches for and automatically selects pixels in the images with shading similar to that of the pixels that were "depth sprayed" by the artists.

"It's a very fast process," says Yewdall. However, there is a creative aspect as well, even though the software automatically calculates a large percentage of the depth information for a scene based on the selective points chosen by the artists. "You lose all the original 3D depth information whenever the scene is transferred to a 2D format. Therefore, our artists must make certain determinations, aside from using the grayscale depth map," he adds.

Once the artists were satisfied with the 3D appearance of the two key frames, they used DDD's DepthTweener tool, which automatically assigns depth information to the imagery in the intermediate frames based on information "learned" from the depth information already assigned to the imagery by the DDD graphic artists. "During this process, we haven't changed the original 2D content," explains Yewdall. "Therefore, the segment can still be viewed in 2D format."