Skip to content
Events
Demuxed
Oct. 29-30 ⏤ London, UK
Inter BEE
Nov. 19-21 ⏤ Chiba, Japan
Resellers
Latest News
MainConcept Completes Management Buyout to Become Independent Company
MainConcept Releases Codec SDK 16.0 Featuring AI-Powered Video Quality Metric VMAF-E and Workflow Innovations
Stay up to date with our latest news and
subscribe to our newsletter >>
vScore

VMAF-E

Rethinking Video Quality for Live and On-Demand.
eva-featured

MainConcept Easy Video API (EVA)

Control hardware and software codecs through a single API.
cmaf-whitepaper

CMAF: Low-Latency at Scale

Watching live event is all about the moment. Delays can be frustrating. This paper introduces the options, highlighting what sets CMAF apart.
violett-blur

VMAF-E

Real-Time, AI-Powered, and Live-Ready

Rethinking Video Quality for Live and On-Demand

VMAF-E, developed by MainConcept and part of the vScore video quality metrics suite, is an AI-powered breakthrough in video quality measurement. Building on Netflix’s widely-used VMAF metric, VMAF-E brings perceptual video analysis into live production, unlocking instant, actionable insights for content creators, broadcasters and streaming platforms.

Learn more:


VMAF-E is available in the MainConcept HEVC Encoder released with SDK 16.0.

Why VMAF-E?

Conventional video quality tools were designed for file-based workflows: slow, fragmented, and never intended for real-time feedback. VMAF-E changes that.

AI-Accelerated Speed

Up to 10x faster than traditional VMAF, delivering immediate scoring for both live and on-demand content.

Live-Ready Accuracy

Performance closely matches standard VMAF, but is optimized for dynamic, time-sensitive streams.

Zero-Latency Insight

Built into the encoding process, VMAF-E analyzes in coding order, not presentation order, so your results are truly real-time.

Trusted Results

Benchmarked against PSNR and VMAF-CUDA , VMAF-E’s scores consistently align within similar levels.

Engineered for maximum flexibility and efficiency

Ideal for Live Streaming, great for VOD too

Get summary scores at the session’s end, or per-frame results as encoding happens, perfect for monitoring or in-depth analysis.

Smarter quality control

Loop VMAF-E data directly into your workflow for real-time optimization, letting you adjust encoding on the fly.

CPU and Workflow-Friendly

Reduced processing requirements free up CPU resources for encoding, instead of after-the-fact analysis.

See a Tangible Impact for Live and VOD

autolive-streaming
Live Streaming

Standard VMAF is too resource intensive for live use. Operators need to know if stream quality is holding up now, not later. With VMAF-E, instantly assess, optimize, and track key metrics frame-by-frame. No more waiting, no more cumbersome syncing.

vvc-decoder
Video on Demand

VMAF for VOD often requires decoding source and output, perfect frame sync, and re-processing for every content edit or filter. With VMAF-E, direct, inline measurement keeps analysis perfectly aligned, eliminates redundant steps, and simplifies even the most customized post-production pipeline.

Technical Highlights

01

AI-Powered Estimation

Neural network delivers perceptual scoring, approximating VMAF to within ±2 points for robust decision-making.

02

Session and Frame Logging

Choose between instant one-shot results or granular per-frame records for deeper insights.

03

Industry-First for HEVC

VMAF-E is MainConcept’s first production-ready AI metric, a testament to our commitment to smarter, more accessible video analytics.

violett-blur

Want to learn more?

Whether you’re focused on live events or on-demand delivery, VMAF-E gives you quality data in real time, unlocking new options for innovation and efficiency.