# VitMatte Matte Refiner in Flame **URL:** \ **Category:** Between Renders\ **Tags:** ml\ **Created:** [September 20, 2026, 8:19pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936 "2026-09-20T20:19:18Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![allklier](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/allklier/32/4309_2.png) [@allklier](https://forum.logik.tv/u/allklier)\ **Post date:** [September 20, 2026, 8:19pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/1 "2026-09-20T20:19:18Z")
An often used technique to improve ML mattes is to run them through an edge refiner model. A popular being VitMatte. It needs a trimap of the edge and will then refine the matte further. Here’s an .inf model for VitMatte, and Flame setup to use it. In my testing I’ve found though that AutoMatte actually has a better result w/o VitMatte, meaning it doesn’t seem to have any benefit. To the contrary actually. That’s a big kudos to AutoMatte then. But feel free to test it, YMMV… And you might find use in other mattes that didn’t originate with AutoMatte. ![image](https://us1.discourse-cdn.com/flex020/uploads/thedepartmentofexternalservices/original/2X/a/a99c8e6624b948e61d9a04e598ea0a00c95d6f48.png) ![image](https://us1.discourse-cdn.com/flex020/uploads/thedepartmentofexternalservices/original/2X/9/9faaa2bc5fb008c8324bdd7221cbaad54cf0db99.png) > **[Dropbox](https://www.dropbox.com/scl/fi/jjskzbfnmbe2j8emnzhni/VitMatte_PE.zip?rlkey=kwgv6opx0gzfnymtaxrlvw4uh&st=rcq9cr6i&dl=0)** ---
**Author:** ![Nils](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/nils/32/12295_2.png) [@Nils](https://forum.logik.tv/u/Nils)\ **Post date:** [September 20, 2026, 11:24pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/2 "2026-09-20T23:24:23Z")
Yeah I’ve worked at Nuke places running SAM then VitMatte, but it’s usually not as good as MatteAnyone(2) let alone VideoMaMa. They left it behind when I showed them SammieRoto. I often start with the whole frame via Sammie then patch in bits where Automatte is better afterwards. ---
**Author:** ![philm](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/philm/32/9286_2.png) [@philm](https://forum.logik.tv/u/philm)\ **Post date:** [September 21, 2026, 2:38am UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/3 "2026-09-21T02:38:43Z")
I’m a big fan of silhouette. BorisFX are really doing outstanding things these days. I’m hoping for FBX or alembic output from their head tracker. If they can do that then it’s trivially obvious that they can do projection receiving geometry too (which is truly something that was always in the realm of the possible for flame). But subtitles Am I right? Onwards. ---
**Author:** ![cnoellert](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/cnoellert/32/559_2.png) [@cnoellert](https://forum.logik.tv/u/cnoellert)\ **Post date:** [September 21, 2026, 3:21am UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/4 "2026-09-21T03:21:06Z")
They’re already doing alembic output for Mocha so it should be trivial. ---
**Author:** ![johnt](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/johnt/32/14040_2.png) [@johnt](https://forum.logik.tv/u/johnt)\ **Post date:** [September 21, 2026, 5:21am UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/5 "2026-09-21T05:21:10Z")
I forgot about the alembic output. I have used it to track hair - unwrap paint. It was useful. Talking of Roto, I have been quite impressed by Onyx AI Roto. Much cleaner than mocha and silhouette AI Roto IMHO. ---
**Author:** ![PlaceYourBetts](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/placeyourbetts/32/6852_2.png) [@PlaceYourBetts](https://forum.logik.tv/u/PlaceYourBetts)\ **Post date:** [September 21, 2026, 6:38am UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/6 "2026-09-21T06:38:21Z")
Yeah. I have been impressed by Onyx. Using it within Nuke all the time ---
**Author:** ![allklier](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/allklier/32/4309_2.png) [@allklier](https://forum.logik.tv/u/allklier)\ **Post date:** [September 21, 2026, 6:53am UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/7 "2026-09-21T06:53:18Z")
I used SmartRoto earlier this week again. Doesn’t get any cleaner… Of course, it depends on the complexity of the shapes and movement. Sometimes AI roto can be faster. Always mix and match. Never limit yourself to one way if speed and quality matter. Silhouette wins on any face segmentation roto you may need for color. ---
**Author:** ![philm](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/philm/32/9286_2.png) [@philm](https://forum.logik.tv/u/philm)\ **Post date:** [September 21, 2026, 12:48pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/8 "2026-09-21T12:48:56Z")
> [@cnoellert](#): > > They’re already doing alembic output for Mocha so it should be trivial. I’ll try it today ---
**Author:** ![cnoellert](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/cnoellert/32/559_2.png) [@cnoellert](https://forum.logik.tv/u/cnoellert)\ **Post date:** [September 21, 2026, 2:06pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/9 "2026-09-21T14:06:53Z")
They’re using it for powermesh output. My point was that Boris has that code in place and implementing it elsewhere shouldn’t be a crazy lift. ---
**Author:** ![philm](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/philm/32/9286_2.png) [@philm](https://forum.logik.tv/u/philm)\ **Post date:** [September 21, 2026, 2:52pm UTC](https://forum.logik.tv/t/vitmatte-matte-refiner-in-flame/14936/10 "2026-09-21T14:52:54Z")
It’s only this year that I’ve ever used silhouette and mocha pro. I used the sapphire tools and Syntheyes for many many years.