Foreground Subject Darkness

Hello What tool what method can I use to reclaim this darkened subject? I attempted to use the lasso tool to snag the subject then used the 'shadow/highlight' tool to adjust. Though the details are intact within the subject the results were less than acceptable. Let me know if there's a posted tip/link/ and/or comment that will assist in the effort. Thanks!

Yes, with film pos, this should work beautifully. This technique is something I developed on my own, and its success or failure for you could greatly depend on your scanner software. But it's such a powerful technique, I found it worthwhile to fool around with it until I got great results. I use Silverfast AI (scanning software), which allows various exposures (in Silverfast AI, it's called "Midtone Brightness"). I typically scan 5 times, with exposures set to -100, -50, 0, +50, and +100 naming the scan files by exposure. Then when using Merge to HDR Pro in Photoshop, when manually setting EV, I choose the "EV" option (as opposed to the "Exposure Time" option), with the 0 (normal) exposure set to 9, and each exposure under that -1.5, and each one over that +1.5. -- So -100 EV=6; -50 EV=7.5; 0 EV=9; +50 EV=10.5, and +100 EV=12. This procedure gets GREAT results in constructing a 32-bit HDR image from scans.
An even better method (using the same scans) is to get the plug-in, "HDR Efex Pro 2," which costs $99.95, but has a 15-day free trial period with full functionality. You can see all its amazing capabilities at http://www.niksoftware.com/hdrefexpro/usa/index.php?view=help/videos.shtml
Since getting HDR Efex Pro 2, I don't use Photoshop's Merge to HDR Pro anymore. :+)

Similar Messages

Dark pictures with canon 5d mark 3 and 600ex flash, night indoors in Auto Mode and flash on.

I have purchased a new canon 5d mark 3 and speedlight 600ex-rt flash. While testing it indoor at night , the room lights are on and I was using the Auto Mode with Flash on. The pictures are Dark with the Flash on and when the flash is turned off the pictures are brighter. I am confused. How to get the AUto mode working with the Flash on.

Automatic mode, it it's attempt to be reasonably foolproof, is going to lock you out of having any control over the exposure.
The E-TTL metering system on your camera & flash are attempting to find the correct exposure for your subject.
Due to laws of physics (the inverse-square law), light has "fall off". It's not actually possible to use a single light source and expose a large room with that single light. Either the subject is nicely lit but everything else will probably be too dark. OR... the background is nicely lit but the subject is heavily over-exposed. (BTW, there are two ways to get around this.)
The amount of fall-off is based on the relative distance from the light. Every time the distance increases by a factor of 1.4 the amount of light is cut in half. This means if you have subject 10' away, but you have a foreground object only 7' away and a background object 14' away, and the flash is exposed for the subject at 10', then the foreground object will be over-exposed (with twice as much light as needed) and the background will be under-exposed (with half as much libght as needed.) Incidentally... if you "double" or "halve" the distance then the light difference will be 4x or 1/4 (depending on if you are getting closer or farther.)
Here's a video which both explains and then later demonstrates the concept in a way that makes it very obvious to see and easy to understand. If it seems confusing at first... stick with it. The video is only 12 minutes long.
In fully automatic mode the camera wont let you control anything (except how far away you stand from your subject).
If you switch to "Program" mode, it essentially works like automatic mode EXCEPT it will actually allow you to override settings or set compensations.
When you use a flash in a room with some ambient lighting (but otherwise dimly lit) there's a technique called 'dragging the shutter' (which the camera can actually do for you somewhat automatically).
To do that, set the camera to Tv mode (Tv = Time value... which is shutter priority mode). Dial in a slow-ish shutter speed... say around 1/60th of a second. Use a moderately boosted ISO (400 would probably work well... you might even go up higher).
The camera is going to evaluate the exposure needed for the room as though you don't have a flash... and it'll set the camera settings accordingly. However... since it knows you have a flash, it'll still perform the E-TTL II exposure metering and will use the flash anyway. What you end up with is a shot that has a nicely exposed primary subject (because the camera used flash) AND... a nicely exposed background which was primarily illuminated with ambient light -- but looks pretty good because the camera used an exposure that was adequate for the ambient light.
The reason I suggest using Tv mode and NOT Av mode is because you can control the mixture of flash vs. ambient light by controlling the TIME that the shutter is open. But this does NOT work if you control either the aperture or ISO.
Here's why: The flash is "momentary" but the ambient is constant. If you were to adjust the aperture settings, that will increase or decrease ALL the light sources that contribute to the exposure... so it increases (or decreases) the amount of flash AND the amount of ambient light ... and it does them in the same proportions. If you want to bright up the background lighting WITHOUT over-exposing your foreground subject, then you don't want to increase both... you only want to increase the amount of ambient light.
Since the flash is a momentary burst of light, if you leave shutter open longer you wont actually collect any more light from the flash (because the flash was providing illumination for only a tiny fraction of a second. Extending the amount of time that the shutter remains open means the camera will _only_ collect more of the ambient light... and that helps bring up the dark backgrounds.
One way to get around the flash fall-off problem is to greatly increase the distance between flash and subject. You still get "fall off" but now the distance you have to change to notice the fall off really increases. But this method usually isn't very practical because you'll be standing much farther then you'd prefer to be to get the shot.
The second way (and much more practical) is to "drag the shutter" (deliberately use longer shutter times so that the camera can collect more ambient light long after the flash is done.).
Tim Campbell
5D II, 5D III, 60Da

Faking rack focus: how to blur a rotoscoped foreground element

I've successfully employed the Roto Brush to cut out a foreground subject, built a grayscale precomp of the scene, and used Lens Blur to simulate a rack focus from the background of the scene to the foreground. After rotoing the foreground subject, I copied the Roto Brush effect to a pure white solid, so that it represents the utmost foreground. The effect is pretty convincing (if I do say so myself), but there's a problem...
START:
END:
The end of the rack focus looks great--foreground subject is nice and sharp--but the problem is that the start of the rack focus looks really fake, like the subject is in the Witness Protection Program. What I'm trying to achieve is the "bloom" that would be present around the perimeter the subject if it were out of focus optically. Instead, it's just sharply cut out, and clearly looks like it's comped over the background.
I've tried "negative choking" the roto on the white foreground solid and feathering the edges, but that really doesn't work--it just creates an in-focus outline around the subject once the rack is complete. I'm guessing what I have to do is comp the foreground subject in and blur it separately, so that the perimeter goes fuzzy. The Lens Blur is currently applied to single instance of the video clip, and is referencing a precomp of the grayscale depth map.
Any thoughts on the best way to achieve the look I'm after? Thanks!

Yes, AE's lens blur sucks, if I may say so. You would have better luck with other tools. Try the Box Blur or a third-party plug-in. Specular bloom can be simulated with a Glow preceding the blur to boost the highlights, light wrap using again blurs, mattes and the Add blending mode. Just don't overdo....
Mylenium

Subject transparent w/Blue Screen ??

I've been using Adobe CS5 now for a few months, and, of course, there is a lot still to learn.
When doing blue-screen compositing, I have tried several exposure techniques on the screen; over exposing one stop, underexposing one stop, and exposing it exactly the same as the subject in front of it. It appears to work best if it's over exposed by one stop.
To composite the front and rear elements is pretty easy and straight forward with the Blue-screen Key, but I'm running into a problem that I just can't seem to solve. No matter what I do, I can't seem to cancel out a slight transparency of the subject 100%. I can always detect a little bleed-thru of the background. I have tried viewing the foreground subject with the Mask Only option, and have made it completely white, and yet it still bleeds through when the mask is turned off again. I have tried everything with the controls, until I'm blue-in-the-face....ha, ha, ha .
Using Ultra Key instead of the Blue-Screen Key does not seem to help either.
Could it be something weird, like the blue-screen material is the wrong color? When I bought it, it was supposed to be for digital video, but if it is for film how much of a difference would that make? Or, could it be something about the background scene I'm trying to composite into the shot?
Any help would be greatly appreciated.
Thanks.

> @ Todd: seems to me you mentioned you were working on tutorial or
something that explored the use of multiple instances of a keying effect
to get a more refined matte. Am I recalling incorrectly?
The "Color keying" tutorial on this DVD is exactly that:
https://www.video2brain.com/en/products-27.htm
I show how to use two instances of Keylight---one for the body and one for the head/hair---as well as a garbage matte and a couple of hold-back mattes for the eyes and the buttons on the subject shirt.
If you don't want to get the DVD, here is the same course online:
http://www.creativeedge.com/9780132616430
I try to not use this forum to actively advertise the non-free stuff that I make, so I hadn't mentioned this video before.
For even better keying results, I like to use a somewhat more advanced technique that Mark Christiansen describes in his After Effects CS5 Studio Techniques book: He uses a super-tight garbage matte created with either manual rotoscoping or Roto Brush (or even Auto-trace), plus a super-tight hold-out matte, leaving a narrow edge in between; and then the keying effect is only applied to that narrow edge.

Can you do multiple ultra keys

If greenscreen has 3 or 4 shades of green can you open a 2nd ultra key in cs5.5 to get the darker dhade of green that was missed the first time around?? and another and another

You should - if you didnt' already - consider also using garbage mattes to cut unnecessary material out of your shots. If there are areas of the screen that never come in contact with the foreground/subject, then you can eliminate some of the compositing work by masking those out with the garbage matte. You can also animate the points on the garbage matte as well so that is an additional layer of control when you're trying to reduce the number of color variances on the background/chroma layer.

Speedlight 430ex ii on a 6D - aperture blinking!!???

Hoping someone can help me understand what is going on here... I'm still learning about using my flash so maybe expecting the wrong thing.... I have my camera set to Tv, and auto ISO, the lights are low in the living room and i take a shot of my sleeping black and white dog. The camera beeps to say it is in focus. Camera takes the shot at 1/125, f/4, ISO 8000. Picture somes out a bit dark. I then attach my speedlight... try to take the same shot and now the aperture is flashing??? ISO tries to use 400 so that makes sense - since it's underexposed ... so i then set the ISO myself to 8000 or higher and the Aperture is still flashing (which i know means that the exposure is not correct as far as the camera is concerned). Why is it happy with the scene without the flash but when i add the flash it's not? I tried a lot of shutter speeds (much slower) in combination with higher ISOs (still in Tv proority) and stilll i have a flashing aperture. I would have thought that since the flash is set to ETTL the camera recognizes that it is there and will provide the necessary light??? is this not the case? If i ingore the flashing aperture (which by the way is always reading f/4 for this shot) and bounce the flash the shot comes out well. if i aim directly at the dog it way over exposes (as you'd expect). Why is the aperture flashing when the flash is on/attached but not when it's off?
Body: Canon 6D, Canon T1i, Canon Elan II,
Glass: Canon 24-70mm f/2.8L II USM, Canon 18-55 kit lens, Sigma 70-300mm f4-5.6, Canon 50mm f/1.8,
Flash: Canon Speedlite 430ex ii

The only modes where the camera behaves differently because a flash is present is in full auto mode and also the Program mode.
In program mode, if you switch the flash off and meter the shot in low-ish light and note the exposure, you may be wide-open with a fairly low shutter speed (set a deliberately low ISO so you can see this... e.g. as I'm doing it while writing this reply, with no flash my camera wants to use f/4 and 1/15th sec.) If you switch the flash ON and re-meter, you get a lightning bolt icon on the back LCD screen (so the camera recognizes the flash) and while the aperture stays the same... the shutter speed bumps up to 1/60th.
In every OTHER mode... the camera behaves as if there is no flash and uses the flash as if it's a "fill" light. BTW... you WANT the flash to do this.
Here's why...
Flash (and frankly any light) follows a light fall-off rule called the "inverse square law".
The best tutorial I've ever come across on this is Adorama's Mark Wallace: https://www.youtube.com/watch?v=nk9cTa3UthM
Assuming you watched that... suppose you are taking a photo of a friend who is standing 7' in front of your camera and flash. Suppose we point the flash straight ahead (even though we could get better lighting using other techniques -- we'll keep this basic) The amount of flash power needed to properly expose a subject at 7' (whatever that is) will continue to spread apart as it gets farther from the camera ... that's what light does... it spreads out. Every time the distance increase based on the inverse of the square of 2 (in other words 1.41 ... which is roughly the square root of 2) the amount of light reaching that distance will be HALF as bright. 7 x 1.41 is about 10'. So your 7' friend is nicely illuminated... but anything just 3' farther away is getting HALF as much light. If you double the distance (14') it's half again... or ONE QUARTER of the total light. At 20' it's half as much again... or 1/8th as much light. So this is why you get a bright foreground subject and the brackground looks hideously underexposed. There are several options to fix this, but one of the easiest ways to is to "drag the shutter".
"Dragging the shutter" simply means that you deliberately leave the shutter open LONGER than needed considering you are using a flash. This allows the camera to expose for the AMBIENT light, but punch the subject with flash to "freeze" that subject. The result is a beautifully will illuminated subject... and yet a nice collection of the available light illuminating the background making for a nicely balanced shot from front to back.
This happens almost automatically because the camera naturally wants to meter for available light EVEN THOUGH it's going to use flash. It will back down the power of the flash to keep from over-exposing your subject.
If the shutter speed drops too much (very dark) then you can get blur caused by camera shake and that wont look good. For this reason, I usually switch to manual mode so I can keep the shutter from being too slow (yes... the computer in the camera may warn you that it's under-exposed, but it will automatically set the flash power based on the E-TTL metering calculation). You want to collect as much "available" light as you can without going so long that you get blur from camera shake.
Tim Campbell
5D II, 5D III, 60Da

SMOOTH MOTION IN FILM INDUSTRY ALTHOUGH SHOOTING IN 24P......

Hello guys!
I always had a general question about film making...
How professional Hollywood films can shoot their footages with such a smoth motion (without any jitters, strobing, blury e.t.c. artifacts) although they are shooting in 24p. 24p is not good especially when we have to deal with panning shots. Interlaced footage is much better in terms of smooth motion. Yet the films are perfect!! So the question is how they can achieve better and smoother motion although they shoot in progressive than in interlaced.
Thanks.

How professional Hollywood films can shoot their footages with such a smoth motion (without any jitters, strobing, blury e.t.c. artifacts) although they are shooting in 24p. 24p is not good especially when we have to deal with panning shots. Interlaced footage is much better in terms of smooth motion. Yet the films are perfect!! So the question is how they can achieve better and smoother motion although they shoot in progressive than in interlaced.
To answer your question directly, there are a couple of things that Hollywood does that you don't normally see when shooting on a smaller crew, smaller budget, etc...
For one, you need to achieve the much discussed shallow depth of field (DOF) over/above what a typical 24p video camera can do. This is a balance of lens focal distance and image sensor size. An old fashioned 35mm film camera has a "sensor" that is many times larger than an HD video sensor. The focal distance of the lenses is also generally a good big larger. Therefore, shallow DOF.
You also need to frame the shot and move the frame just like in Hollywood. Even when you're looking at a large panoramic vista shot, you'll notice that the image pans only very slightly, and many times the subject/foreground remains relatively stationary. This reduces stuttering substantially, to the point it's almost unnoticable (though you WILL notice it if you are actively looking for it). For tighter shots in motion (say, a full-body profile shot of the subject walking across a room), you're going to be panning or trucking at a speed that will cause strobing in the shot more than likely, and this only mitigated by keeping the subject spatially stationary in the frame throughout the move.
Another technique has to do with lighting, combined with the two previous techniques. If the relative luminosity of the foreground/subject is substantially greater than the background or other elements in the frame, the strobing is less noticeable on those secondary elements. Your eye becomes more focused on the foreground brighter objects - similar to the way DOF affects your perception - and so you just don't really notice that dark, strobing background so much.
The 24p strobing effect has very little to do with processing at all. However, converting between frame rates and formats CAN sometimes exacerbate all manner of video artifacts, including strobing in the shot. It's not the usual culprit though.
So hopefully you learned a new word today - TECHNIQUE - and you better understand that it's not as much Hollywood's technology that aids the quality of their product, as it is the TECHNIQUE involved in the production. That should be good news of course....technique can be learned, whereas much everything else about Hollywood's production aesthetic comes with an enormous price tag that most independent producers cannot afford.

Copying an alpha channel from one comp to another?

I've created a green screen key using Primatte in one comp, and I want to use that same key in another comp, but where the background has since been replaced and rendered out with an effect I want to keep on the foreground subject. I bet it's something simple...
Actually I bet that explanantion's confusing.
Subject X is against green in a video in comp A.
X was extracted using a key and rendered with an effect and a new background to creat video B.
I want to extract subject X from B (and keep the new effect) using the alpha channel created in comp A.
So how do I copy and paste the alpha channel from comp A to a new comp with video B in it?
Cheers!

Pre-compose, use as a matte layer.
Mylenium

Need help placing people / objects into another photo ...realistically

Hi, I've been a Photoshop user for many years now and have adopted CS4 when it initially came available. My photoshop skills are to a good standard however one thing I always find difficult is taking a selection from one photo (usually a person / object) and placing it into another photo.
I am able to correct the size and scale and maintain the aspect ratio but I never seem to be able to make the photo look convincing, I know this is a basic part of Photoshop, but I was wondering if any of you could offer some tips or advice on how to make the person / object more realistic and hopefully so that clients won't be able to spot it was missing in the first place.
Thanks in advance

The Daily show is a satirical show in the US with John Stewart. As a kind of running joke on "Photoshopped", they often badly photoshop things together, cut and paste style.
Whether things will ever look correct, is to do with LIGHT. The light on your background has to be the same as the light on your foreground subjects. Color can be easily tweaked and matched but light cannot. Photographers with the right expertise (thinner on the ground than you might think) can look at the light in a destination image, and try and match in the studio, but this is a highly skilled area. I once had a professional job where someone had taken David Beckham to a studio - and they wanted the images realistically comped into beach scenes, with broad sunshine, resulting hard shadows, sunsets and horizons. Unfortunately the photographer was obviously more interested in having fun with a high profile client like Beckham for their book, than they were in lighting correcty for the intended destination images. Impossible without weeks of painting, and I told them. No point in wasting clients time on images like this.

Need advice working with Metal-FX files...

I've just been given a large job and at the last minute the spec of the job was changed to the Metal-FX process, and the files provided to me have already been converted, meaning they are 4C+spot channel (silver). On most of the images I need to remove the silver spot layer from underneath the foreground subject.
Some of the simpler images will be easy to work with, but a couple of them require the compositing of 3 and 4 other images together for the final image. Is there any easy way to do this and bring this spot channel into layers for easy editing and blending? It doesn't seem easy from what I've tested. It seems that I will need to possibly work the 4C in layers with alpha channels and multiple silver channels which will need to be merged together using the alpha channels from the 4C layers.
Or am I missing something here?
Thanks for any info you can provide since I've worked only once or twice using spot colours about 5 years ago.
Chris

Seems like you are missing the point of Metal-FX, its designed so that the silver does not need to be knocked out by the cmyk.
All the files Layers will have 5 colours you don't need to think in terms of the silver needing to be a separate layer. If you need to knock out the silver make your selection then use the channels panel to edit just the silver channel.
Metal-FX works best if you avoid solids (100%) in the cmyk colours.

Template Questions

I have a question about templates for my site to display my
artwork.
I would like to have a top level template with some text
across the top of the screen - these would be years: 2006 2005
2004...
If the user clicks on one of these years, I would like a
second line of text (according to the year) to appear: paintings,
drawings, photographs... Different years will have different
subjects. Each of the years would be created from a nested
template.
The user could then click on painting or drawing and a series
of thumbnails would pop up.
My problem: I would like to use templates so that I could add
years later without having to add it in each page of the site - but
I would also like the viewer to have some idea of which year and
subject they are viewing - so that when they come to the site,
click 2006, then paintings, they page will have 2006 and paintings
slightly darker than the other years and subjects to remind them of
where they are. Is this impossible with templates? It seems like if
I have each year link to a separate page with a darkened year that
I won't be able to use a top level template at all... Should I use
frames instead?
Here is what I am working with (I plan to reverse the
locations of the subjects and years) - as you can see, there is no
way to know, just by looking, what year and subject are chosen:
http://www.carlosferguson.com/new%20site/Pages/Paintings/Paintings%202006%20Group%201.htm
Thank you very kindly for any help!
Carlos

> Wouldn't I have to go through each
> page individually and add that year to each?
Not if the year display were either part of the template's
non-editable
region, or were part of a server-side include.
> I can't make one of the years or subjects darker than
the
> others (as these areas have to be un-editable regions in
order to take
> advantage of the template's possibilities.
Sure you can.
It's easy to do with CSS or with javascript.
With CSS you would have a little stylesheet embedded in the
head of each
page that allows you to set the down state for any given
button (each button
would have a unique id - for example -
<a href="foo.html" id="button1">...<a
href="bar.html" id="button2">
and then your stylesheet would have this -
a#button1 { styles }
on one page, and on another this -
a#button2 { styles }
See how that works?
With javascript, put this in script tags in the head of the
document -
function P7_downImage() {
var g7="<imagename>"
var g7url="<pathname>"
if ((g7=MM_findObj(g7))!=null) {g7.src=g7url;}
and this on the <body> tag
onload="P7_downImage()"
Then on each page you would need to make two edits:
Set g7 to the *name* of the button (not the file name but the
HTML name -
e.g., "productsbutt" - without the brackets), and g7url to
the pathname to
the button (e.g.,
"images/nav/button3.gif" - without the brackets), and bada
bing, bada boom!
There is an excellent tutorial here -
http://www.projectseven.com/support/answers.asp?id=126
Murray --- ICQ 71997575
Adobe Community Expert
(If you *MUST* email me, don't LAUGH when you do so!)
==================
http://www.dreamweavermx-templates.com
- Template Triage!
http://www.projectseven.com/go
- DW FAQs, Tutorials & Resources
http://www.dwfaq.com - DW FAQs,
Tutorials & Resources
http://www.macromedia.com/support/search/
- Macromedia (MM) Technotes
==================
"jcferguson" <[email protected]> wrote in
message
news:[email protected]...
> thanks for the reply:
>
> I am trying to creat a site with a top level that the
user could click ( a
> year for example ) that would then bring up a second
level of
> possibilities
> ("subjects") for that year.
>
> Once the user has chosen a year, I would like that year
to be darker in
> color
> than the other years on that top level.
>
> Once the user has then chosen a subject, I would like
both the chosen year
> and
> the chosen subject to be darker in color than the other
possibilities, so
> that
> the user had a sense of where they were in the site.
>
> I can do this with all individual pages, but what if I
want to add a year
> later (say..., next year, for example). Wouldn't I have
to go through
> each
> page individually and add that year to each? If I try to
do this with
> templates though, I can't make one of the years or
subjects darker than
> the
> others (as these areas have to be un-editable regions in
order to take
> advantage of the template's possibilities.
>
> I have a feeling this is a simple answer, but a
difficult to phrase
> question.
>
> Thanks,
>
> Carlos
>

Advice on best workflow for large Motion project

I am a part-time video editor/designer/motion graphics creator/etc. Normally, I work on projects with pieces no longer than 5 minutes, even if the projects themselves might be 30-40 minutes of total material--mostly video support for conferences and awards shows.
Right now I am embarking upon a mark larger project--10 30-minute segments, each of which is 100% motion graphics. They all involve a speaker against a green screen for the entire segment with the motion graphics keyed in front of and behind him.
We recorded this directly to hard drive in a studio that had a VT4 (Video Toaster) system, so the best Mac-compatible codec they could provide me for clean green-screening was full-resolution component video. This is giving me great keys, but I also have about 500 GB of raw footage.
In this project, I need to first edit all the takes from each episode into a clean 30-minute piece, and then add the motion graphics. And this is where my question comes in. It seems to me FCP is much better for editing the raw video, but that Motion is where I want to do just about everything else. I need to somehow bring the video into Motion, because I want to create "real" shadows against my background from my keyed footage.
When working with a long project, and with a full-resolution codec, what is my smartest workflow? I am trying to spend the least time possible rendering back and forth, and also avoid generating huge in-between files each step of the way. It seems that any way to approach it has plusses and minuses, so I want to hear from people who have been there which path gets me to my goal with the least hassle.

I need to somehow bring the video into Motion, because I want to create "real" shadows against my
background from my keyed footage.
"Real shadows are only faked in Motion. You have many options including a simple drop shadow or a copy of your matte layer filled with a gradient and a gradient blur applied with a distortion filter so it appears to be projected onto the wall. Be sure to take the time to make this a template effect and to keyframe the shadow angle if the foreground subject moves.
When working with a long project, and with a full-resolution codec, what is my smartest workflow? I
am trying to spend the least time possible rendering back and forth, and also avoid generating huge
in-between files each step of the way. It seems that any way to approach it has plusses and minuses,
so I want to hear from people who have been there which path gets me to my goal with the least
hassle.
Well, you've got two conflicting interests. One, you have to sync the Motion work with the video of the keyed speaker and, two, you have to edit. But it seems to me that your planning must include lots of design work up front, media you can re-use or modify slightly, text formatting that can be precomped, a large stock of effects you will apply over and over again. Do all of this stuff first.
You also want to explore working at lower rez through your planning and roughing stages. for instance, there's no reason to pull a full rez copy of your foreground into Motion if all you need to do is sync to his audio and get rough positioning. You can put him over black and export all of his clips using any medium to low rez codec at reduced frame rates and just use the Screen Blend Mode to drop him roughly onto your Motion projects.
You'll get lots of advice over the next few days. If you're posting to other Motion or motion graphics forums, please do us all a favor and return someday to all of your threads and tell us what you did and what you learned.
bogiesan

Questions about divulging information about future features

Good morning,
I'm making my way through this thread for the first time.
As someone who has passed the California Bar and who has been a stockbroker in the past, I would like to comment on the statement of Steve Forde, who was the Senior Product Manager for Visual Effects for Adobe for portions of 2011.
Steve said,
"Unfortunately - we get stuck in a legal pinch about commenting on ANYTHING that may or may not be in future products, off the record or otherwise."
I'm asking that you consult your internal SENIOR legal counsel again regarding this statement that you assert prohibits you from giving any details whatsoever about what the Adobe Team may or may not aspire to having in future versions of After Effects.
I'm asking this of you for a couple of reasons.
Firstly, no attorney denies that the law changes over time and recent court decisions have made shareholder derivative lawsuits much less popular as a way for the plaintiff's bar to make a quick buck by extorting money from companies who may have made aspirational statements to shareholders or others in conference calls or elsewhere.
As I understand the current trends in Delaware Law, (which, as the situs for quite a few corporations governs large swaths of their conduct) as long as appropriate disclaimers are clearly communicated at the time of the aspirational statement, such language is now permissible.
Such standard disclaimers could include language such as
"we cannot guarantee whether any feature discussed will appear within one year or within five years or whether it will ever appear at all. It is currently a goal of the design team to reintroduce this feature, but we reserve the right, in our exclusive discretion to change that decision for any reason whatsoever, without consequence.
Additionally, you are vested with no rights to rely on any assertion or representation made based on any such aspirational statement and we insist and demand you consult an attorney before embarking on any course of conduct, such as purchasing stock or altering any current or projected business plan, based in any manner upon what we say here.
However, given the above caveats, it is the current goal of the Adobe design team to reintroduce this feature within the next three releases of AE again, subject to change at our sole and exclusive direction in any manner, time or mode of our choosing, without prior notice of any kind."
If you had a standard disclaimer such as the above that you attached to any and all such statements (by having it appear within the signature block of any Adobe employee posting to this forum), it could take conversations such as this to a more rewarding level for both Adobe staff (who could share their very profound dreams and aspirations with us) and the Adobe community at large, which provides inspiration and helpful feedback to Adobe employees seeking to assist us.
Then the focus could shift to whether or not you were divulging trade secrets. But it would still permit you to be more informative in a general way.
So I ask that you run this by your SENIOR legal counsel again. Remember, just as there can be different ways to tackle the same 3D graphics animation challenge, each with rewarding results, attorneys similarly can vary from attorney to attorney and new and bright ideas on how to approach such issues do occur from time to time, especially in light of the massive favorable swings in case law supporting the rights of massive global corporations such as Adobe viz a viz the rights of the little guy.
Sincerely, Matt Dubuque
Here is Steve's quote in its entirety, from May 26, 2011:
Unfortunately - we get stuck in a legal pinch about commenting on ANYTHING that may or may not be in future products, off the record or otherwise.
That being said, as Michael explained in a previous post, Vector Paint got stuck in the conversion of AE to 64 bit. It's something we hear quite a bit from our users that Vector Paint is important. Again - wish I could give you more than that, but we are bound by very strict rules.
Therefore - it is important to us, and the more we hear via this form the more likely it is to get resources behind it in a future release. Again I realize it is impersonal, but the results of this form are sent via email to a real human every time someone hits submit; then that person brings the results of submission to a weekly meeting on the AE team where we discuss features and scheduling.
Steve
Sr. Product Manager - Visual Effects
Adobe Systems Inc.

Attaching an appropriate disclaimer may be fine for an attorney to dot all the i's and cross all the t's, but there is one thing it will NOT do: it won't stop people from misinterpreting what they read.
Most people go by what they PERCEIVE to be true and not the ACTUAL truth, a fact that people in marketing like me deal with all the time.
Steve Forde might write the following totally hypothetical and totally fictional line in a blog:
"We're thinking about a feature for a future version. Right now we call it the Clean Plate Generator. You would be able to shoot a single clip, panning and tilting the camera as desired, and After Effects would automatically create an image whose total size matches the totality of the image shot in the clip. It would even remove selected foreground subjects such as a stray person who inadvertantly walks into the shot."
Sounds great, yes? And with your proposed disclaimer, all would be good; Adobe would have covered its corporate pink posterior. Or would it?
Do you wish to speculate how many people would read that as a statement of a definitely-going-to-be-there feature in the next AE version?
Can you imagine the can of worms that would open when this fictional feature isn't in the next version?
Would it really be worth the massive and expensive fence-mending campaign Adobe would have to undertake among angry users who were erroneously looking forward to this fictional feature?
All of which could have been avoided simply by not saying a word?
I can see where attorneys may like a hypothetical situation as the one above: they could charge more billable hours! I doubt Adobe would like it, however.
Legalese is one thing, maintaining happy customers is another, and there is no amount of legalese that can overcome people's nasty tendancy to perceive what they read as what they WANT to read instead of what is actually written.
If people would read what was actually written, there would be far fewer posts in this forum that say in essence, "Hey, how come it takes two weeks to render ray tracing on the brand-new HP student laptop I just got from Wal-Mart?"

Tweaking edge in keying with Keylight

Hi, I'm having trouble getting rid of a wiggling fur around my foreground subject after keying with Keylight 1.2.See screenshots in status and final result views (when I preview the outer edge comes alive in a horrible wiggle).
I've tried tweaking with Screen Gain, Screen Matte white and black, Screen Shrink/Grow, Softness etc. I just can't seem to get rid of this wiggling halo. I admit my background bluescreen wasn't perfect, but I should still be able to get a better result that what I'm getting. Can anyone help? Cheers

What is the source of the footage? You may simply be seeing compression artifacts which should be removed before keying. Your 2nd image is still queued (use the camera indstead of an attachemnt top avoid that), but that would be my first guess. If so, you may need to work on that. This can be done with plug-ins like Magic Bullet Frames, Revision FX SmoothKit or techniques inside AE. A simple one would be to use Shift Channels to extract the luminance, blur it slightly, then add a duplicate of your layer in Color blending mode on top of it to recombine it. All of that in a pre-comp, of course. There are many other ways, but give it a spin and see if it makes a difference. In the effect itself you may need to adjust the Despill Bias (decoupled from the Alpha Bias) and the Clip Rollback. This usually helps.
Mylenium

GS Workflow with Color

I'm looking to see if anyone has found a seamless way to correct green screen footage using Color. I did some test sequences with GS on V2 keyed over a BG on V1 and sent it to Color. The GS footage comes through unkeyed. I tried nesting the 2 clips with the same results. If I proceed in this fashion, the corrections will change the shade of the green BG and possibly mess up the key when sent back to FCP. If I send the sequence to Color first, I may make the key more difficult to pull. Since there is more control over color corrections in Color, I'd like to be able to use it to correct and match the GS and BG. It seems the only way I figured out how efficiently use Color with GS is to key the footage in FCP, do basic corrections and matching to the foreground and background using the filters in FCP. Then export that section as a QT movie, bring it back into FCP and drop in place in the timeline, then send the sequence to Color for final grading. There has to be a better way.

Andy Matthias wrote:
There has to be a better way.
There is. But it's not easier.
You need to treat the key and the fill separately.
Use the uncorrected footage to pull the key and generate the alpha signal.
Then also send the footage through Color - with the object of correcting for the subject and neutralizing all the green. IOW, make the green go gray and make your foreground subject look great.
Render it out.
In FCP use the key you pulled and fill it with the rendered footage from Color.
The big problem here is that you can't tell in Color if you've been too aggressive in removing green and are bleeding into the foreground subject.
This workflow is a bit more intuitive in FCP where you can see how everything interacts. Or even better - After Effects which is much more polished for this sort of thing.
HTH
- pi

Foreground Subject Darkness

Similar Messages

Maybe you are looking for