MPEG 153 summarized by streaming standards emissary Mohamad Raad

MPEG, the organization made up of international multimedia standards-setting working groups, wrapped up its 153rd meeting on January 23, 2026. Unified Streaming’s standardization representative Mohamad Raad attended the online-only meeting.
Over five days, working groups focused on three main subjects: media authenticity, DASH, and common encryption.
Following are Raad’s MPEG 153 highlights.
Media authenticity
In a world where AI slop’s run amok, verification of the provenance, or origin, of media has taken on a high level of importance.
One method of checking media authenticity is deploying a (non-MPEG) specification called C2PA, short for Coalition for Content Provenance and Authenticity.
So what’s the news on the spec?
Well, version 2.3 of the C2PA specification, now out, features a new section for handling live video, to which Unified Streaming researchers contributed in a significant way.
“Keep in mind C2PA is also the basis of JPEG Trust,” said Raad. “And (JPEG Trust) has had a second edition approved for publication as an international standard this month as well,” said Raad.
According to the JPEG site’s overview of the framework, JPEG Trust (ISO/IEC 21617) seeks to establish trust in media, including “aspects of authenticity, provenance, attribution, intellectual property rights, and integrity through secure and reliable annotation of the media assets throughout their life cycle.”
The C2PA spec continues to grow, covering more media types and more use cases, said Raad.
Common encryption
Theft of media streams is common, so encryption better be just as common. Luckily there is a format called Common Encryption (CENC).
“There has been significant progress understanding the threats” to common encryption, said Raad. Likewise, MPEG has also made similar progress realizing “potential solutions to meet those threats.”
DASH
DASH-related work is coming back to life.
With the delay of the publication of the sixth edition, though, developing text for an amendment has been slow-going, Raad noted. There is some effort to define extensions to ISOBMFF to meet the authenticity requirements. That is also progressing slowly.
Gauss you had to be there
On the compression side, Gaussian splat-based coding is maturing as a topic as well, said Raad.
Gaussian splatting is a volume rendering technique that reconstructs or visualizes photos or videos as photorealistic, three-dimensional scenes. Just FYI.
Four MPEG 153 questions for Raad
What do you think is the most important industry issue that MPEG 153 addressed?
It was ‘what’s next for MPEG,’ really. MPEG has had a great deal of success with some past standards (CMAF, ISOBMFF, DASH, AVC, AAC, etc). Although the discussion was held in the systems working group (WG) this time around, I think this applies equally well to the other WGs as well.
Favorite thing about this latest MPEG gathering? What got you excited?
It was good to hear experts’ thoughts on the future of MPEG systems and what they feel needs to change for MPEG to deliver high-quality standards in the years to come. On the technology side of things, I didn’t see any major developments in the major MPEG themes during this meeting.
You know the saying. If the MPEG will not come to Mohamad, then Mohamad will go to the MPEG. What do you think MPEG 154, held late April 2026 in Santa Eularia, Ibiza, Spain, will focus on?
The groundwork for the next video compression standard will continue. We will see more AI in the proposed or demonstrated technologies in all the compression fields.
Gaussian splat coding will probably move into point cloud- and audio-related standards, rather than being a general solution for video.
MPEG will continue to make small incremental steps in the area of media authenticity. I also expect to see a more mature common encryption-related work item.
One work item that could become more significant with time is the blockchain for media monetization activity. Although it’s not well attended at the moment, it may actually provide a very useful solution to content monetization and protection.
Blockchain? So we’ll be paying bitcoin to watch cat videos?
Yes, it’s the smart contracts activity within MPEG systems which may provide the basis for solutions to monetize content in a standardized way. Smart contracts are blockchain based, providing techniques for tracing the origin of content as well as the usage and consumption of that content. With the correct provenance metadata in place, these smart contracts are intended to use that metadata to determine who owns what, and how they can be compensated for the use of their content.