Print Page
Saving Bandwidth Before and After Compression
5/22/2014

When it comes to preserving bandwidth for multichannel distribution of video, much of the discussion has centered on new compression algorithms (H.264 AVC and H.265 HEVC) that reduce bits and maintain pristine image quality. That's because less data enables faster transfers, less storage, and lower operation costs while providing the opportunity to add more streams and channels for higher revenue.

However several content providers are working with technology suppliers to develop workflows that process content before (Faroudja) and after (Cinova) the compression stage. This, they say, reduces bandwidth even further by identifying redundant bits and making adjustments in software accordingly. The perhaps best part, this technology is agnostic of compression standards, so it works with all existing infrastructures, so wholesale equipment upgrades (set top boxes and other decoding devices) are not necessary when migrating to the latest compression tools.

Faroudja Enterprises, a rebranded company located in Los Altos, California that specializes in video processing, has introduce its Video Bitrate Reducer(VBR100) patent-pending technology that is employed at the front or back end of a video processing and video over IP delivery workflow, before or after compression, and, according to the company, reduces video bit rates from 35 to 50 percent; without any perceptual loss of quality. And it is interoperable with any type of video compression or resolution, from standard definition to 4k.

The new Faroudja Enterprises' technology also provides a concurrent lower resolution version at no extra bandwidth cost. For example, additional benefit (before and after compression is applied) is gained with 4K sources by delivering a 1080p parallel feed at no extra bandwidth cost, the company said. Company founders include Image enhancement pioneer Yves Faroudja, his wife Isabell and Dr. Xu Dong. The founders said they have realized that their experience in video technology is applicable to video compression.

The company said it designs and markets pre-processors and post-processors, used before and after compression coding/decoding, to achieve a lower bit rate and better image quality with existing codecs. It does not perform compression coding or decoding, but instead designs pre-processors to be used after decoding at the consumption device.

Faroudja Enterprises' technology gives users either a choice of better image quality at the same bit rate, or if required, a 35 – 50 percent bit rate reduction for the same image quality as the original. In either case, compression artifacts are reduced and/or eliminated. They said significant results have already been achieved—two patents have been granted, and three more are in the works. The processes are suited to a wide range of video, from teleconferencing and videophones, to standard definition (SD), high definition (HD) and Ultra High Definition (UHDTV) applications.

The company said that digital video compression systems' efficiency often can be improved through the use of a Faroudja pre-processor (prior to compression) and a post-processor (after compression decoding). The workflow might include the use of a support layer in parallel with the conventional compression path (see diagram). The Faroudja scheme complements conventional compression standards (MPEG-2, MPEG-4, and HEVC) and does not require any modifications of the standard codec.

Results are accomplished, Faroudja said, though the Faroudja support layer that helps provide full resolution at reduced bit rates. It can also be configured as a transcoder to help convert video between existing compression formats, such as MPEG-2, H.264, etc., to and from Faroudja Enterprises' Support Layer format to save bandwidth, bit rate, or file size in the cloud without sacrificing image quality.

"Demand for video bandwidth is doubling every 3 years, yet network compression schemes efficiency doubles every 10 years," said company founder Yves Faroudja. "It is clear that a fundamental change must be made in how networks operate. Our new technology provides a solution for this today, easily integrates into existing systems, and is fully compatible with future compression schemes (e.g. HEVC, VP9, etc.)."

Cinova, located in Mountain View, Calif., is another company that thinks it can improve upon current encoding schemes by working with a file after it has been compressed. The company, founded three years ago by CEO Dr. Anurag Mendhekar (who worked at the Xerox Park technology think tank and Yahoo), using a proprietary technique it calls "Perception Optimized Processing" as the basis for a software product called "Crunch."

Mendhekar said they have come up with a set of parameters that describe the human visual system. They apply these parameters to every macroblock of ever frame of an encoded video. Taking the output from the encoder, the software looks at every macroblock (4x4, 8x8, etc.) of a frame and applies the parameters to that macroblock. From the results Crunch can compute a visual sensitivity index. Based on that index, the user can then decide how aggressively (or not) to transform that macroblock further or throw away unwanted elements of that macroblock. The result is improved picture quality and bandwidth saving of 20-50 percent, Mendhekar said.

"We came at it from the angle that videos are viewed by human beings, so let's take a more explicitly human visual systems-based approach," Mendhekar said. "What we've come up with are parameters that we feel accurately describe the human visual system. We reduce bitrates by up to 50 percent and we guarantee not to negatively harm to a frame of video. If I wanted to increase the savings even more, I could say, 'let's adjust the video quality to get where you need to be in terms of bit rate.

"Today if a content provider wants to set 6 Mbps as a 'house" format, there's probably a dozen different parameters on their encoders to get to that 6 MBps," he said "If they know that Crunch is coming after the encoder, they may reset the operating parameters on the encoder to a higher bit arte, knowing they'll get the bandwidth savings at the other end with Crunch. This ensures good picture quality."

Starz, a U.S.-based premium television company with 17 cable /satellite/telco TV-delivered channels, is testing Crunch at its Meridian, Colorado headquarter facility for video over IP delivery. With Crunch, in the company's R&D lab, Starz has internally tested distribution tiers with bit rates reduced by up to 45 percent. Lower bit rates mean less time downloading a video for the consumer and there are fewer chances for the video to stall.

Using a 37-minute test video that includes all types of compression-challenging content from different Starz delivered shows. Starz, with its current infrastructure, takes 2 hours to produce IP distribution packages on an Intel i7 processor-based workstation that has 24 cores (two processors, each with 12 cores.). So, it took 48 core-hours to process an IP distribution package. Processing the same material with Crunch, using the same bit rate Starz currently uses, tests indicate it may require as little as five core hours.

"We have only evaluated it in our lab," said Felix Nemirovsky, Video Systems Architect, Starz, adding that there are no current plans to deploy it live. "We have utilized it in a number of ways but, all encode tests to date were for video over the Internet applications."

"In some tests we used 4 MP4 files of difference resolution and produced 4 Crunched versions of those files," Nemirovsky said, "thereby totaling 16 streams at various bit rates, again for sets of adaptive bit rates to be delivered to multiple type of Internet devices."

The reasons for Starz's interest in Crunch are 1) Bit rate reduction on a per movie feature bases rather than the same transcode setting for all movies; and 2) More efficient production of GOP aligned streams for ABR applications, than a transcoder, but with the requirement of delivering the same resolution as the original source.

"We are not focused on bandwidth reduction because we are not providing bandwidth," Nemirovsky said. "We are focused on higher quality deliver at current bit rates."

Cinova's Mendhekar said aside from bandwidth savings and maintaining quality, using Crunch can reduce the number of transcodes a content distributor needs from eight to four, while maintaining the capability they already have.

One of Cinova's main messages is that Crunch can deliver the exact gains that HEVC (H.265) provides to deliver video over IP networks.

"We think HEVC, and its promised 50 percent bandwidth reduction, will be widely adopted in five years at least," Mendhekar said. "We'll we're delivering that kind of results today. Instead of ripping out your H.264 encoders, I can give you the 20-50 percent promised with HEVC, today, without changing out your infrastructure."
  Print Page