Adobe leads high-quality raw video format initiative

Effort called CinemaDNG tries to set standard for digital video file formats. Also at NAB 2008, Adobe adds speech transcription to video metadata.

Adobe Systems thinks we can do better with the quality of digital video images. It is also developing a way to search on the audio within video clips.

At the National Association of Broadcasters (NAB) Show 2008 in Las Vegas this week, Adobe will announce a joint initiative to develop a specification that it hopes will eventually lead to a file format for higher image quality.

Adobe will show a preview of technology that will create a text transcription of the audio within a video clip at editing time. CNET Networks

The effort is called CinemaDNG, named after the DNG (Digital Negative) raw digital still image format designed by Adobe. The company is working with others in the industry including camera makers and software developers, said Simon Hayhurst, senior product manager for dynamic media at Adobe.

The group's hope is to have a specification ready sometime this year and to submit it to a standards body to encourage broader industry adoption, he said.

Initially, the specification will only affect "high-end Hollywood and top-end indie" filmmakers because equipment that supports this format would be the most sophisticated and expensive available. But eventually, this format could be used more broadly.

"It lays the foundation for the correct way that you want to do cinema in the future," said Hayhurst.

Creating a common standard will help accelerate adoption of higher quality imaging, he said.

The advantage of the specification will not only be better resolution, but it will also give more image control to cinematographers and editors. The format can be useful for archiving films which could be reissued with a different look as well.

Adobe intends to support the format in future versions of its video work-flow products, like After Effects and Premiere Pro.

"You want enough space to innovate but have commonality so that you are implementing technology when there is a genuine need for it to be different," Hayhurst said.

Video to text
Separately, Adobe will give a preview at NAB 2008 of technology that automatically transcribes the audio track of a video file.

For editors, this will allow them to more quickly find passages within a clip based on a text read-out of the audio. The output of the video-editing software will also include that transcribed information.

As a result, viewers of a Web video will be able to search on terms to find a specific location within a video.

For example, a person could search a CNET video review for a product name and a specific feature, such as camera zoom.

Adobe will demonstrate the feature on a version of its Soundbooth audio-editing product under development and on Premiere Pro.

The company intends to support the feature in the next major release of its video work-flow software. There was a two-year gap between the releases of Creative Suite 3 and 2, so the next major version is likely to come some time in 2009.

The transcription information will be stored in XMP (Extensible Metadata Platform), another format developed by Adobe.

"We keep saying that metadata is the most important thing happening in our industry and we want to prove it," said Hayhurst.

In other announcements, Adobe will announce that it is now natively supporting Sony's XDCAM EX tapeless video file format in its Creative Suite 3 video-editing tools.

And it is adding support for H.264 standard, high-definition video format on its Flash encoding software. It added support for H.264 for Flash video playback last year.

 

Join the discussion

Conversation powered by Livefyre

Show Comments Hide Comments