Question about processing video
Posted: Tue Sep 06, 2011 7:00 am
Hello.
I've few little questions.
I'm wondering something. I'll describe what i used to do for denoising.
- I'm building a noise profile using several part of what i think being the most representive (it may, but not necesseraly be the most important) noise level.
For movies, i'm building one profile for each.
For series, i'm building one profile sampling on several episodes, and process the whole serie with it.
I'll consider building 2 models for series going over than 5 years, when state of masters of 1rst episodes may be more "old/damaged" than last.
Can't afford building 100 profiles to have one/episode...
After, i'm processing the whole serie or movie with this builded profile.
Do i make a big mistake or is it relatively good ?
Is there any interest of activating "adaptative" in my way of processing ?
If yes :
- Will this in best case only provide improvement and in worst not improve, or can it in worst case produce less good results ?
- Will it slow down process ? (No, a little, medium, a lot).
Is there any benefit to rebuild noise with v3 profiles i've made with v2.2/2.6 ?
Thanks.
I've few little questions.
I'm wondering something. I'll describe what i used to do for denoising.
- I'm building a noise profile using several part of what i think being the most representive (it may, but not necesseraly be the most important) noise level.
For movies, i'm building one profile for each.
For series, i'm building one profile sampling on several episodes, and process the whole serie with it.
I'll consider building 2 models for series going over than 5 years, when state of masters of 1rst episodes may be more "old/damaged" than last.
Can't afford building 100 profiles to have one/episode...
After, i'm processing the whole serie or movie with this builded profile.
Do i make a big mistake or is it relatively good ?
Is there any interest of activating "adaptative" in my way of processing ?
If yes :
- Will this in best case only provide improvement and in worst not improve, or can it in worst case produce less good results ?
- Will it slow down process ? (No, a little, medium, a lot).
Is there any benefit to rebuild noise with v3 profiles i've made with v2.2/2.6 ?
Thanks.