How Would You Do This???
JoeMamma2000
Posts: 2,615
In a couple of recent threads a link was posted to a very well done, 4 minute video made, in part, with Carrara.
And I began to take a closer look at the video, pretty much stepping thru it, to get a sense for what it took, in practical terms, to accomplish the entire production. And the closer I looked, the more impressed I became with the detail and difficulty involved in the production.
So I thought it might be instructive to extract just one frame and have everyone (who's interested) try to figure out exactly what was required to get that frame.
There are many shots with a similar difficulty, but I thought this one, which is part of a 3 second shot, is especially interesting.
Now, the image below is the frame, and I'll list some of the requirements that had to be met to accomplish the shot:
1. There are concurrent, complicated animated graphics on a CG screen
2. There is a live action, over-the-shoulder shot of a hand, lit appropriately for the surroundings
3. There is a live action reflection shot of an actor's head, lit appropriately for the surroundings, in the CG control panel
4. The motion of the hand and fingers had to match the animated graphics on the CG screen (ie, the graphics had to respond to the hand movements and light up as the fingers touched the screen, and the hand had to initially pause as the CG "screen" became active)
5. The motion of the reflected head had to match the motion of the hand, and the context of a person sitting down at the control panel just seconds before.
6. There is a white glow in the bottom right hand corner simulating the glow from environment lighting.
So the challenge is to figure out the workflow, hardware/software requirements, etc. Basically, how the heck would you do it?
If you want to view the context, I believe it's about 52 seconds into the video.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
lol! I will reply to you Joe
not with answer as to how THEY did it but with how I would do it.
I do not know how they did it
I would render a spherical scene of the cockpit whatever as a sequenced image series with crew movement
use it in the background with the scene tweener oscillated so it looped
I do this myself a lot.
the screen on the consol would also be an image sequence, the slider at a set keyframed point until the finger touched the button the slid along to where the action was needed
I have done this too using a character by a laptop touching the keyboard.
the appropriate lighting can be achieved using global illumination with the animated background
this I have not and am unlikely to do as I want to get stuff done quickly and quick it would not be.
I suppose I better actually watch this video!
Dartanbeck linked it for us in the positive thread.
yes, I have seen it before and the behind the scenes one http://vimeo.com/m/17761334
the greenscreening obviously took a lot of work but they use shadowcatch models of the components seen in the chroma footage and indeed with that as an image series in the near background on a plane using transparency which can be done readily in most greenscreen programs even cinegobs keyer which I use albeit badly against my blue kitchen wall, and is indeed visible in the textured view in Carrara for compositing.
After Effects or similar compositing program. There are no 3D models in this frame and no reason to use Carrara or *any* 3D program.
8-/ next bogus challange please.
It's probably After Effects. In the comments below the post on Vimeo he tells what software was used and it's on the list.
Congratulations, Joe. Now you can back up your claim that it wasn't all done in Carrara.
(I know, I'm supposed to be ignoring you, but I was curious.)
First thing I'd do is to fix the GLARING mistake they made and replace that yucky human with a Proper Cow.
Everyone knows humans are out of fashion these days.
Okay, kids, it's not some childish attempt to back up a claim that is so obvious that anyone who knows even the slightest about film production would accept it as being ridiculously obvious...
And no, I'm not playing the childish games that so many here thrive on.
This is a legitimate attempt to get folks here to THINK and to LEARN about the stuff they are doing. But only here in this forum is that considered a bad thing, and viewed with skepticism and ridiculous childish game-playing. Only here do you have to drag people, kicking and screaming, to actually learn something.
Anyway, if there are any here who aren't interested in playground games (though I've really started to wonder if there are), here's some things to consider:
!. Yes, of course it's done in After Effects. It's called compositing. Where you take a bunch of layered graphics and superimpose them on each other to make them appear that they are a single shot.
2. So what they did is to shoot either one or two actors (one over-the-shoulder shot of the hand, and a separate front-on video of just his face), using an HD video camera on a tripod, in a studio, with a light setup (called a "softlight") and composite that with a CG shot of the control panel.
3. And in order to be able to composite the layers together, the live action camera footage had do be shot in front of a "greenscreen".
Now, there's a lot more to it than that, but from that starting point, let's see if we can figure out the STEPS, not just the software, that YOU or I would have to perform to make that entire shot.
And if you just want to play childish games of calling me names and discrediting me, I've heard it all before. And as Kodiak likes to say, "get off my thread".
Oh, and one thing you can figure out is how the CG part was done in Carrara. Because it was done very well, and it wasn't easy.
And if you really want to learn something, look at this 3 second shot in the video and see if you can reproduce it using Carrara. And instead of the live action, use an M4 or whatever. But do it in multiple shots, like they did, and composite it together. It will be VERY instructive, and get you thinking about the process and what's involved. Though it will be a WHOLE lot easier than what the producers had to do using live action footage.
And in case anyone cares, the whole field of compositing is, IMO, fascinating and probably one of the funnest parts of visual effects. And if you really want to have fun, pick up a book on compositing to see how it's done.
And if you really want to twist your brain, consider this...
Before they started making that video they had to plan this and all the other shots. So they had to decide:
"Okay, I want a shot of a guy sitting down at a control panel, the panel starts up, and he does some hand motions to activate the projection camera.
Now, how are we gonna do that?
Hmm.....well, we don't want to build a control panel in the studio, so we'll make it CG, and use Carrara or whatever. But we want a real hand. Which means some live action footage. Which we'll have to composite. Anyone know After Effects?
And wait, to do that we'll need a greenscreen studio...and a camera....
And oh, BTW, we want the control panel to look cool, so you should be able to see the reflection of the operator in the panel. How the heck are we gonna have a live action shot appear as a reflection in a CG control panel???
And the controls have to respond to the operator. So if we also have some video of a hand, the hand video and the head reflection video need to match, and all the motions need to be in sync with the CG control panel.
Hey, does anyone know how to make a real cool control panel in Carrara? And it has to have animated graphics. Hmm....
Oh, and the lighting. First we have to design the interior of the craft to see how the lighting should look so we can make live action lighting that will match...
Hey, where we gonna find a studio with all this stuff?"
Very interesting, and very complicated....and that's just for ONE 3 second shot.
Enough with the talk. Show us a render of what you've done on this so far, JM2000.
Okay, so I raise the issue. I spend my time (a lot of it) writing up explanations and discussion. I challenge others to do it as a learning exercise.
And you expect me to also spend my time working on renders for you too?
Would you like me to get you a cup of coffee while you wait?
Incredible.
Okay, so I raise the issue. I spend my time (a lot of it) writing up explanations and discussion. I challenge others to do it as a learning exercise.
And you expect me to also spend my time working on renders for you too?
Would you like me to get you a cup of coffee while you wait?
Incredible.
When I started my Space scene thread, I was willing to post examples. In fact, I had intended to all along. It's not an unreasonable request.
I agree that compositing is fascinating. It's also a huge time saver. I'm having issues with bandwidth so I can't review the video now, but I did see it before. I think there would be better 3D shots to choose from than one with fairly simple 2D motion graphics. How about the guy walking down the corridor? maybe the moon terrain with the shadow crossing over the old lunar rover? Maybe the ship positioning itself in lunar orbit? Maybe the dolly shot of the camera moving in on the audience?
This isn't from the video obviously. It's something I'm working on. The sample image is four layers: The stars, the black hole, the ship and the robot. The shot is brief, and not that complicated, but it saved a lot of render time rendering in layers and then compositing.
I've also just rendered a couple spherical shots so that I can get faster renders yet. Much quicker to wrap a 2D image around your scene than the heavy geometry and complex shaders of the model I'm working with.
When compositing video with 3D footage, one normally uses Z-pass (or called Depth Pass). After Effects and other compositing programs do this. After you color correct the video footage and 3D footage, you would want to add a Z-pass from the 3D program or you can do this in After effects.
Then one would adjust the Luma so one can adjust the chromatics of the green screen to the 3D render. Typically, the luma would be a greyscale or even black and white.
What that does is it gives the compositor the ability to make the Camera video green screen footage appear as it part of the 3D footage and vice versa.
The reflection trick is not that difficult to duplicate and is done quite often in architectural renderings. The architect wants to show reflections of existing buildings in the windows and surfaces of the 3D composite. So the architect would make a Z or Depth pass for the 3D composite and integrate it as described above onto the 3D structure or one can make the 3D structure reflect the luminance of the actual "real" buildings.
Luma tricks, color corrections and z-passes are often how one can integrate fire, explosions or otherwise impossible 'survivable' situations into live video footage and make it appear as if it was part of the actual shot.
You really didn't think those scenes in Avatar with all those winged creatures and spaceships with people sitting on or in them while flying around was real now did you?
EDIT: Added a quick show and tell (no specific details - there are tutorials and classes online about ZDepth and Luma - Mattes)
Pretend the top 2D drawing is the 3D model. I use the very basic (not detailed) steps below to show how to add the video to the 3D rendering.
You- You- You mean Unobtanium wasn't real?!? ;-) I don't even want to know about Santa!
Okay, so I raise the issue. I spend my time (a lot of it) writing up explanations and discussion. I challenge others to do it as a learning exercise.
And you expect me to also spend my time working on renders for you too?
Would you like me to get you a cup of coffee while you wait?
Incredible.
Yep, 2 sugars in it please. So what if it takes you a week or two to put it together, then 4 or 5 days to render? Do you have anything better to do? Aren't you just a hobbyist like most of us?
"....fairly simple 2D motion graphics...."
"The reflection trick is not that difficult to duplicate..."
:gulp:
Which is why there are so many of you guys producing such beautiful, professional videos like this one. Maybe you guys can team up and reproduce the clip as I suggested to show us how it's done. Simple...not that difficult...should be a piece of cake for experts like you.
How 'bout by tomorrow? Cool. Thanks.
I edited my post to show how it is basically done.
If you want to learn After Effects and compositing, take some classes online or at a university.
Video Co-Pilot, Creative Cow and other places that are After Effect specialists offer many online courses (both free and paid.)
If you look up z-pass, luma matte, color correction and etc for After Effects, I'm quite sure you'll find many examples to show you how it can be done.
Actually Joe, I'm doing you a favor by getting you started in the right direction. I don't have the time nor the inclination to keep you entertained.
You're a smart guy, I'm sure you can find what I told you is accurate and have a play for yourself; assuming that is, you have a compositing program capable of such tasks.
OK. Like everyone has said, this is NOT how they did it.
How would I do it and use Carrara in the process? Take into consideration that this is NOT what I do so I'm just pulling stuff out of thin air.
1. The live action would be shot with a green screen so we could get an alpha channel thing so we could lay the animated stuff beneath it. I've never done anything like that but I think that's they way the do it. So that removes the live action of the hand.
At the same time, I'd get a separate head shot and save that down as an AVI.
2. I'd create the "control pad" stuff in Photoshop. This would be several different versions of the panel with an all dark version, buttons lit versions.
3. I'd place a plane (or some sort of surface) in Carrara at the same angle as we want the control panel.
4. To simulate key clicks or whatever, I would automate the control pad texture by adding keyframes where the the texture would change from one control pad version to the appropriate next keypad version.
5. Finally, to add the reflection, I would add a second layer on the shader for the control pad. This would overlay the control pad stuff and would be the AVI of the live action head.
At least, that's how I'd try to do it. I didn't go back and watch the video again but I think that covers the basics.
because a crappy video I made a while back in Daz studio with little effort has gone viral and I had to unsubscribe from Youtube emails to stop clogging my inbox today.
6. Don't forget the blue light sphere blob in the lower right corner... foggy light spheres in a sterile dust-free environment are like mandatory in space.
And Mr. Design Acrobat, I do appreciate it. Truly. From the bottom of my heart.
Because, if I can merely, just for a moment, bask in the brilliance that is you, then I'm happy.
If I can just scramble for some tiny scraps of the vast expertise you might occasionally toss in my direction, then I'm happy.
I know I'm not deserving of any of your knowledge, your wisdom, or your many, many decades of training and experience as a true 3D Master. So I am thrilled to receive any of your insight.
Because you, Sir, are truly a Design Acrobat.
Thank you.
And Mr. Design Acrobat, I do appreciate it. Truly. From the bottom of my heart.
Because, if I can merely, just for a moment, bask in the brilliance that is you, then I'm happy.
If I can just scramble for some tiny scraps of the vast expertise you might occasionally toss in my direction, then I'm happy.
I know I'm not deserving of any of your knowledge, your wisdom, or your many, many decades of training and experience as a true 3D Master. So I am thrilled to receive any of your insight.
Because you, Sir, are truly a Design Acrobat.
Thank you.
Tsk,Tsk
If you don't want to learn from me, then just say so. I offer you what I know, accept it or don't - it's your choice.
As far as my expertise, I have little as I only started with 3D graphics in 2006 and 2D graphics a few years earlier. I depend upon others that are much more knowledgeable than I. My Uncle, for example, worked as a film editor at MGM studios (now retired) and I pick his brain when I can. I visit with my game designing buddies as much as I can and as much as they have time for me to pester them with questions. I watch online live feeds of experts who show how to do things and take questions from their viewers. I enroll in courses online as much as I can afford them so I can learn. As mentioned earlier, Video Co-Pilot and Creative Cow are just super places to visit.
One thing I have learned in over six decades of life is that you always learn something by watching others and, if the opportunity arises, ask them questions. I'm an engineer by training and don't have natural artistic ability. What I've learned about computer graphics comes slow and most of the time I struggle with the concepts and of course, the implementation.
I don't come here much to the Carrara forums or display what I do as I'm working with Lightwave3D and there is a forum for that, so I go there. I also belong to a private forum of like-minded individuals (rank amateurs, but with a mixture of their own specialties as programmers, game artists and etc.) who have fun prodding and poking at each other as well as experimentation in design challenges or generally how they are doing with their graphics or life in general.
Anyway, it's too bad you are so skeptical about me and people in general.
6. Don't forget the blue light sphere blob in the lower right corner... foggy light spheres in a sterile dust-free environment are like mandatory in space.
Dagnabbit.
I knew I left out the most important part.
Fine. And I started in visual effects (actually started in photography for a couple of years, then into film and vfx/3D/2D) back in 1980. And in fact I may even know your uncle (or at least know the name), depending on when he retired. I've been doing this stuff for a very long time.
So while I appreciate that you're willing to teach me, I already know the answer to the challenge I posted. I posted the challenge in an effort to get others to THINK and LEARN and maybe broaden their perspectives about the craft, instead of arrogantly asserting an expertise which doesn't really exist.
Instead, I get put-downs and name calling and arrogant attitudes from those who want to look smart or whatever, but in fact who couldn't even begin to duplicate what I posted.
I posted it as a learning exercise. But if people want to act, as usual, like they already know it all, and couldn't be bothered to learn about anything else, then fine. You'll fit in fine with all the other self-assumed experts here.
But if anyone is TRULY open to learning something new, I strongly suggest you take on the exercise and try to duplicate the video clip.
Fine. And I started in visual effects (actually started in photography for a couple of years, then into film and vfx/3D/2D) back in 1980. And in fact I may even know your uncle (or at least know the name), depending on when he retired. I've been doing this stuff for a very long time.
So while I appreciate that you're willing to teach me, I already know the answer to the challenge I posted. I posted the challenge in an effort to get others to THINK and LEARN and maybe broaden their perspectives about the craft, instead of arrogantly asserting an expertise which doesn't really exist.
Instead, I get put-downs and name calling and arrogant attitudes from those who want to look smart or whatever, but in fact who couldn't even begin to duplicate what I posted.
I posted it as a learning exercise. But if people want to act, as usual, like they already know it all, and couldn't be bothered to learn about anything else, then fine. You'll fit in fine with all the other self-assumed experts here.
But if anyone is TRULY open to learning something new, I strongly suggest you take on the exercise and try to duplicate the video clip.
I don't know if you would know my Uncle or not as he is in his 80s now. He worked primarily in TV and did editing spots for TV commercials. For reasons due to age and a couple of strokes I only remember specifically the TV series Daktari as one he worked on. I know he did more, but didn't keep up with what he worked on.
Glad you have a background and perhaps you can share your expertise. When I write on a forum I hope that others besides the one I am responding to will learn something or that others will tell me "I'm on the wrong track". It isn't usually to just one person. The purpose of a forum, I think, is to be like a giant sand box where everyone can have a play.
I doubt I can duplicate the video as I don't own any video equipment other than my cell phone. :)
The 3D I can do as well as the compositing. I'd have to dig up some public domain footage, perhaps some old NASA videos.
So he was with Ivan Tors? I remember Daktari, but I was a kid when it ran. Doubt I'd know any of the editors. A lot of those guys were really pioneers, so I have a ton of respect for them.
This thread can be summarized thus:
Joe - 'Check out this video clip and tell me how they did it!'
Audience - 'We don't know for sure but we can guess. Green screens. There were definitely green screens involved. '
Joe - 'Well that's obvious isn't it?'
Audience - 'We're pretty sure they used Adobe after effects.'
Joe - 'Yes yes but how did they make it dammit?!'
Audience - 'Well, why don't you tell us?'
Joe - 'You're all idiots! How can you learn anything if you don't at least try and reproduce the clip?'
Audience - 'Got a sample of how you did it?'
Joe - 'I'm just here to point out how stupid you all are. I'm not gonna throw anything constructive your way! Fools!'
The end.
So he was with Ivan Tors? I remember Daktari, but I was a kid when it ran. Doubt I'd know any of the editors. A lot of those guys were really pioneers, so I have a ton of respect for them.
No, Mr. Tors died a long time ago.
Perhaps you can show us some of your VFX and photography work so we may be inspired.