Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Infra red problem with clothes

M

Marco Jahn

Member
Joined
Sep 14, 2010
Messages
5
Reaction score
0
Points
0
Hey everybody,

we had a daylight shooting last week and some black clothes are purple.
How can we fix this problem in post. Are there any preferences in redcine X or in storm?

Here is a Screenshot of the problem. The clothes normally are black.
Infra%20red%20problem.png
 
Last edited:
Wow! It looks like you used non HOT-MIRROR ND filters for that one! Is that RED M or RED MX?

There is no setting like the one you are looking for in REDCINE-X or STORM but you could try keying the magenta colour in a compositing software and desaturating/darkening it.

Good luck with that!
 
Interesting that the sunglasses, the phone and the seat are correct black but the fabric seems to emanate IR. Was the phone on the cans black too? She looks like a bad key - look at her right shoulder. Hmmm....
 
Hey Guys,

thanks for the quick response.
We used a normal Tiffen ND 1.2 and a Tiffen ND Soft Edge 0.3. in a Arri MB19.
I don`t know why there is a white edge at her right shoulder.
Yes, the phone on the cans were black too.
What I don`t understand, in another shot we don`t have these problems. Same filters.
Here a Screenshot:
Bildschirmfoto%202011-02-10%20um%2012.48.11.png
 
Last edited:
Its IR pollution. You have to get combo hot mirror ND's or get an IR cut/ mirror filter to put in the front of your stack. Hate to be the the bearer of back news but you will probably need to reshoot this scene if the clothing must be black. Its very difficult if not impossible in many cases to correct IR pollution.
 
See my thread from a couple years ago: http://reduser.net/forum/showthread.php?t=9141

All the answers and explanations you could need are in there. Shooting in the winter time seems to make it much worse. Different fabric reacts very differently to IR also cloud cover and time of day effect it as does some lens coatings.

The only real post fix is a bitch, as mentioned earlier keying out the magenta and reducing it's saturation is it. Be very careful with skin tones while doing that though, you might need. To roto out the faces so ey don't turn to lifeless. Good luck
 
Also Marco I can tell your second shot does contain a ton of IR pollution, its just not as obvious. You will be stunned by the difference when you pass an ir cut filter in from of the lens in a scene like that.
 
Ok, thanks for your reply.
Is it possible to put an IR cut/mirror directly on the lens and use normal ND filters or must the IR filter be put in front of all other filter?
Fortunately there are only a few shot. Now we have a little work to fix it in After Effects.
 
There exists an interesting phenomenon with polyester fabrics.

For many years, now, wedding photographers have been plagued with white wedding dresses made with polyester.
It seems that rather than just reflecting light, the polyester threads REFRACT the light hitting them, and give off more of a cyan tint that really shows up on film and digital image sensors.
The bottom line is that there is a possibility that this is happening in the posted shot in addition to the IR issue, and I fully agree that the IR issue is the primary cause.

Another situation that I ran into during my career doing visual effects is that there are people who see pretty much normal colors, yet are "color blind" to deep purples. Clothing like that in the posted shot may very well appear black in person, yet show up with a strong tint on an image.

Indeed, I remember reading in one of the threads that neither the M or the MX sensor has an infrared blocking filter, and that an external IR blocking filter is very strongly recommended.
 
Last edited:
Mike Prevette said:
See my thread from a couple years ago: http://reduser.net/forum/showthread.php?t=9141

All the answers and explanations you could need are in there. Shooting in the winter time seems to make it much worse. Different fabric reacts very differently to IR also cloud cover and time of day effect it as does some lens coatings.

The only real post fix is a bitch, as mentioned earlier keying out the magenta and reducing it's saturation is it. Be very careful with skin tones while doing that though, you might need. To roto out the faces so ey don't turn to lifeless. Good luck
Click to expand...

Yeah the tone of her fingers and forearms are going to make this very very very difficult to do.

If you can pull it off, I'd love to see it.
 
@ Randall: both M and MX have IR blocking filters. But it's a delicate adjustment to put an IR filter directly in front of the sensor without making it too strong. If you do, you would introduce a green tint, since it eats into the visible red instead of blocking IR only ( no filter is perfect).

ND filters block the whole visible spectrum, but they don't block IR. So, if you use a strong ND to reduce visible light, you shift the balance between visible light and IR to extremes: a faint image versus very strong IR. If you'd try to avoid that from the start, your image would lack red if you were shooting with less ND or without any.

Some Sony cameras have problems with the longest wavelengths of visible red, BTW.
 
Randall Weatherington said:
There exists an interesting phenomenon with polyester fabrics.
Indeed, I remember reading in one of the threads that neither the M or the MX sensor has an infrared blocking filter, and that an external IR blocking filter is very strongly recommended.
Click to expand...

This is true for the M but not for the MX. MX has stronger IR filtration and can go without a hot mirror for any ND below 1.2


Elsie N said:
Pretty sure I remember someone from RED saying the "purple sun" issue has been handled, and I think one of the changes for the M-X sensor was a stronger IR filtration. Since the same sensor will be used on Epic, I suppose that applies for it as well.
Click to expand...

Deanan said:
Correct on both.
Click to expand...
 
Yes, both shots are deeply IR polluted.

The main problem in color correcting them is qualifying the correct regions.

I'd scope out the affected areas and color correct them. Rotoscoping would solve the qualifying issue.

Of course its much more labor intensive.

So, its fixable in post- but it would cost you. The results would be OK for TV, but probably not feature quality- though I might surprise me. Also, I'd be correcting it in a compositing application, not a grading application.

If you can reshoot with the right filtration, that would be my recommendation.

My question is why you didn't see that on set? Or did you see it and decide you couldn't do anything about it?
 
Nylon becomes transparent under infrared light, you're actually seeing through her clothes...
 
I love those Formatt IR+ND filters.

Stephen
 
You must log in or register to reply here.
Back
Top