IRIS: The Extra Set of Eyes Every Colour Pipeline Needs

Colour is one of the last things an audience notices when it's right. And one of the first things they notice when it's wrong.

A LUT applied incorrectly. A transform missed during versioning. A render that doesn't quite match the approved master. A title card that shifts a comparison by a few frames. These things rarely happen because someone doesn't know what they're doing. They happen because modern post-production is genuinely complicated, and the sheer volume of versions moving through a pipeline creates a lot of opportunities for something to drift.

As productions become more distributed across facilities, cloud platforms, editorial teams, VFX vendors and colour departments, maintaining consistency from approved master to final delivery is both more important and more difficult than it's ever been. That's the problem IRIS, origami’s automated colour verification system, was built to solve.

The Problem Nobody Talks About Enough

Here's what's actually happening on most productions right now. Once any tool has created plates, teams are checking all of them, 100%, every time. Even where parts of that process are automated, the reality for most data I/O teams is that someone is loading those plates into an application, lining them up against the offline, and then applying CDLs and LUTs shot by shot across the entire sequence in order to compare them before anything goes to a vendor. It's painstaking work, it takes a long time, and more often than not it happens because LUTs and CDLs were mismatched somewhere earlier in the pipeline.

That's the real problem IRIS solves. With IRIS, 99% of shots are now fully automated end-to-end. The only time a user needs to get involved is that remaining 1%, where there's a genuine issue with the LUTs or CDLs in the workload. That shift removes an enormous burden from data I/O teams and gets them focused on actual problems rather than hours of tedious admin. It's the most important thing to understand about what IRIS does, and it's the thing that tends to land hardest with anyone who's sat through a plate check.

Why Manual Colour Comparison Alone Doesn't Scale

Most colour verification still relies heavily on human review, and it should. Creative judgement belongs with experienced colourists, online editors and QC operators, and that's not changing. But modern productions create an enormous number of opportunities for accidental colour drift: multiple delivery versions, HDR and SDR outputs, localised versions, VFX turnovers and returns, platform-specific masters, last-minute editorial changes, distributed cloud and on-prem workflows.

Every additional handoff is another opportunity for something to change unintentionally. The more versions being delivered, the harder it becomes to manually compare every one against an approved reference. That doesn't mean teams need less QC. It means they need better tools to help focus their attention where it's actually needed.

What IRIS Does

At its simplest, IRIS compares a rendered file against an approved reference and automatically identifies areas where colour values deviate beyond defined tolerances. Instead of relying solely on manual visual checks, it runs an automated verification pass that highlights potential issues for review before delivery. The result is faster confidence checks, earlier detection of discrepancies, fewer manual comparison passes and reduced risk of delivery errors.

Importantly, IRIS is designed to support human review, not replace it. The final decision always sits with the colourist or QC operator. IRIS just makes it easier to know where to look.

Getting Rid of False Positives

One of the biggest challenges with any automated comparison system is avoiding false alarms. Anyone who has done reference matching knows the problem: title cards, slates, leader material, countdown sequences and different programme start points can all trigger alerts that have nothing to do with an actual colour issue.

The latest IRIS update addresses this with configurable reference offsets, which let teams skip a defined section of the reference file before comparison begins. That means comparisons can be aligned directly to programme content rather than introductory material, so you get more accurate results, fewer unnecessary flags and less time spent manually trimming. On episodic content and versioning-heavy projects, that can quietly remove a surprising amount of repetitive work from the delivery process.

Automation That Supports Skilled People

One of the things we talk about a lot at Origami is using automation to handle the repetitive stuff so skilled operators can focus on what they're actually good at. We're not interested in automation for its own sake. We're interested in removing friction from the workflow so our team can spend their time on the creative and technical decisions that genuinely require their expertise.

IRIS fits exactly that philosophy.

Instead of spending hours checking files that are probably already correct, your team can focus on the small percentage of deliveries where something genuinely needs investigation. That's a better use of everyone's time and it means fewer things slip through.

Why Colour Consistency Matters Beyond Finishing

Colour management isn't just a finishing-stage concern. It touches every part of the workflow: editorial reviews, VFX pulls, conforms, online finishing, mastering, versioning, delivery. A colour discrepancy introduced at any point can ripple through the entire process, and the earlier those issues are identified, the easier they are to correct.

This matters especially on productions where media, metadata and colour information are moving between multiple systems and vendors. Origami's wider platform is built around maintaining continuity across all of that. IRIS extends that principle by helping ensure the visual result stays consistent from approved reference to final delivery.

The Colourist Still Makes the Call

There's an important distinction between colour verification and colour grading. IRIS does not decide whether a grade is good. It does not determine creative intent. It does not replace an experienced colourist's eye. What it does is provide an objective verification layer that identifies when a delivered file differs from an approved reference.

Think of it as an intelligent warning system. When a shot drifts, IRIS catches it. When a comparison aligns perfectly, IRIS provides reassurance. When something looks different, the colourist makes the call. That's exactly the relationship between automation and expertise that we think works well in practice.

An Extra Set of Eyes That Never Gets Tired

The reality of post-production right now is that teams are being asked to deliver more content, more versions and faster turnarounds than they were a few years ago. More spreadsheets and more manual checking isn't the answer to that, and expecting talented artists and technicians to spend their days performing repetitive validation tasks doesn't make sense either.

IRIS adds an additional layer of quality assurance without adding additional layers of process. When delivery deadlines are tight and expectations are high, having that extra set of eyes matters. And unlike most human operators, it never gets tired.

If you'd like to know more about how we use IRIS and Origami as part of our colour workflows, get in touch for a demo.