Video compression connects the quality of a visual experience with the practical limits of networks, storage and compute. Its next opportunities extend from content delivery to systems that use video to understand the world.
DISTRIBUTED VISUAL SYSTEMS
Smart infrastructure & mobility
Distributed cameras produce continuous visual streams, while connectivity and retention budgets remain finite.
The opportunity to explore
Explore representations that retain information needed for traffic analysis, asset monitoring and event review, with less unnecessary data movement.
What must be evaluated
Evaluate task accuracy alongside bitrate, temporal continuity and performance in difficult lighting. A smaller stream is useful only if the relevant evidence remains usable.
The following are application hypotheses for research and partner evaluation, not announcements of deployed Deep Refactor products or customers.
EVIDENCE & OUR INTERPRETATION
Three signals shaping the opportunity.
01
The uplink deserves more attention.
Ericsson’s June 2026 report found faster uplink than downlink growth at 43 of 55 measured service providers.[1]
Our view: this strengthens the case for evaluating visual processing closer to capture, especially when applications continuously upload images or video.
An observed provider sample; it does not quantify demand for any particular codec or supplier.
02
Machines change what quality means.
In January 2026, MPEG advanced Video Coding for Machines to Draft International Standard, with machine-task performance central to the compression objective.[2]
Our view: AI application builders need to evaluate the information their models use, alongside what a human viewer sees.
A standards-development milestone, not a claim of final standardization or Deep Refactor certification.
03
The codec roadmap remains open.
ITU reports that JVET issued a joint call for proposals for compression beyond VVC in July 2026.[3]
Our view: there is continuing space for research, careful benchmarking and integration work across established and emerging approaches.
A research programme does not establish commercial readiness or guaranteed efficiency gains.
Read compression claims with their conditions.
ITU’s 2020 VVC summary anticipated about 50% less bitrate than HEVC at equal subjective quality, while separately reporting about 40% on 4K/UHD tests using objective metrics. The distinction matters: content, quality measure and test setup change the result.[4]
From technical progress to commercial value.
We see the most credible opportunities in specific workflows with a measurable constraint: an expensive uplink, a growing video archive, a demanding edge device or an AI task sensitive to visual detail. This is a market thesis, not a revenue forecast.
For technology partners, an evaluation should compare the entire pipeline: transmission and storage savings, encoding and decoding cost, latency, compatibility and downstream AI accuracy. Improvements in one layer can introduce costs in another.
For a UAE research company, the opportunity is to connect this global technical progress with the operating needs of regional application builders: representative evaluation, adaptation and integration. Commercial adoption depends on validated outcomes and fit with existing systems.
ITU-T Study Group 16 · Meeting concluded 3 July 2020
Sources reviewed 15 November 2026. Industry findings are attributed to their authors. Application scenarios and market interpretation are Deep Refactor’s proposed perspective.