Transcripts
1. Motion Tracking with C4D R20 for Beginners: Hey, how's it going? I'm Derek Kirk of to tron and I wanted to create a tutorial that went over a ton of motion tracking, tips and tricks, just best workflows and everything. And I kind of created this weird thing from a scene. Let's check it out. So now I will show you their way of tracking new blood up. I'm gonna show you how I did that, how to do other motion tracking things, some examples, what works, what doesn't work? I really wanted to create something that went over from start to finish the entire project. So I wanted to create a tutorial that showed you how to set up your shot, how to pick your camera error. If you don't have a camera, even you can use your cell phone. How to use that to even get a good track how to shoot it. I'm gonna show you how to edit your footage either in a third party software or within Cinema 4D. How to track it at the 2D trackers, auto and manual. How to 3D solve your camera, and then how to reconstruct your scene with a mesh so that you know exactly where things are in your scene. One thing I definitely wanted to make sure I showed and I had a lot of questions about it was I wanted to show how to composite your 3D objects actually into your footage and how to render that out. And I'm going to use redshift. And with redshift, it's super easy if you don't have redshift. I will also show you how to do it with the physical renderer in cinema for 3D. And then we're probably taken in After Effects, add some post effects onto it and be done with it. So we're gonna go from the very beginning to the very end of this process from motion tracking. That's really all I want. I wanted to present something that was flushed out. It wasn't like here's a small piece of the puzzle. I wanted to present every piece of the puzzle so you could actually start putting it together yourself. Go out there right now with cinema 40, shoot something with your phone if you needed to track it, render it, put it together.
2. How to Film for Motion Tracking: There isn't a easy shoot it push button. It's gonna work perfectly. That's not gonna happen. And there's also no way that I can tell you exactly how to do it every time they get a perfect track every time. But what I can do is tell you what works the best, what doesn't work, what to look for, what to avoid, things like that. And I have done a lot of testing. I went out and shot with different cameras. I showered and expensive camera shot with an osmo pocket. I shot with my phone. Whether you have a camera or anything, I want you to be able to do this. So I went out and I shot with my phone and I've got an old iPhone six s. So it's not even that greatest in NDP. But I can't control what didn't control the shutter speed or anything like that. So as far as motion tracking goes, there's a lot of variables. And what I found out is that what it comes down to the most is how you shoot it and what you shooting. So firstly, things you wanna look for. You've gotta have plenty of light and you've gotta have your exposures gotta be correct. Things are too bright. It's not gonna be able to discern the contrast between details and things. And the point is that's the way the tracker works. It notices patterns and pixel areas and it searches for that nice little boxes it creates. It's like I'm looking for this within this bounding box, right? We'll go over that. But if it's all white, it's not going to know the details that it needs. So motion blur, big thing a lot of people, you know, they understand emotion blur is, but they don't really understand how it works in a camera. So basically your shutter speed control is your motion blur. Shutter speed is how fast the aperture closes and the lens and stuff. This is apparently what that looks like in my head. But basically, the more higher the shutter speed that CRISPR the images, you know, like the motion blur, normally like a filmic motion blur saved filming in 24 frames per second, which is what most film is, then your shutter speed is one over 48. It's double your frame rate is to get that cinematic motion blur. So what that is if you're moving a lot and you pause the frame, it's blurred, right? And it helps things not look fake and I'll look robotic. It's really important. But tracking doesn't like that. It's like I'm looking for Chris blends. I want to be able to say these pixels are what i'm looking for. But when you blur them, it's like wait a minute, those pixels I was just looking at that clean image is now gone. So now I have this blurry image. And it's like, I don't know how to track that same with depth of field if you're trying to do something with a really shallow depth of field. So if this blurry, it's not going to track, it just doesn't know how to read blurry things. So that's, that's also important. Shutter speed also, Chater also affects your depth of field and the focal length of your lens. So shutter is good, higher shutter is good because it'll be. Crispr, so your motion blur won't be as bad. But if you can't control your shutter, doesn't mean you're necessarily going to have heavy motion blur. How you move the camera is going to depend on your motion blur more than than that, you can compensate the not being able to have a high shutter speed because I'm able to have a high shutter. You need to have a lot of light. Because the higher the shutter speed, the less light that's getting into the lens. So you've gotta have a brighter area, which is why a lot of motion tracking stuff you see is outside. The sun is look the best light that exist. The sun is fantastic. You can crank your shutter speed up to the thousands and it looks perfect, right? But inside, if you don't have like a studio setup and lights, the actually control your scene. You're going to have a tough time getting stuff unless you have a really good lens. So lower your shutter speed allows more options for you. You get to shoot inside with that a lot of light you can shoot in places that you normally could issue with, without the equipment that you would need if you wanted to do a high shutter speed. So in order to do good motion tracking, without high shutter speed, with motion blur, the way you move the camera is going to be the most important. Now I did a lot of tests. I tried real shaky like Blair Witch Project shaky doesn't like that. You know, walking like running desert, really like that either. But, you know, whip pans know anything that's going to cause like a lot of motion blur. Avoid that. But if you just take it and use Hold it, you're gonna get natural shake in your lens depending on what it is, right? So but if you just walk, if you handle it steady and just kinda do this or like maybe you can do like a slow pan, tried to do it as smooth as you can. But if it's handheld, it's going to be shaky and it's gonna make it a better track. But try to walk slowly. Pretend you're in band practice, right? Like you've got a stabilizer is something like you walk heel to toe, heel to toe. And so you're kinda just slowly going, trying to be really steady. Don't force the shakiness, like that's gonna happen naturally if your hand holding it, okay? Even if you have a gamble, gambles aren't perfect either. And it's okay if it's not super shaky. If you own a dolly, that's fine. You're gonna make sure your move creates parallax, and that's what I want to get into what you're seeing now and talk about Parallax. Parallax is when you have an object in the foreground and object in the background. And as your camera moves, these to move accordingly within your shot. So it knows well this one's moving faster to the left while this was going slower, that means this is closer to the lens. So it knows, oh, these objects are in different space. So it's like, well, this is the foreground, that's the background. So it's going to calculate distance that way. Saying with up and down that can thing. So you've gotta make sure you have stuff in your scene. If you just try to track a wall, like just a wall, it's. It's not really going to know without anything in the foregone. Like that's why you can't just put your camera on a tripod and like pan, it doesn't know without moving, it doesn't know that what you're looking at isn't just a mural like a picture, like a psych wall kinda thing without things actually changing in space, in relationship to each other. It doesn't know that there's depth there. So it's gonna, you're gonna come back and only track points are gonna be really flat. And it's not going to know. So we need to look for you need to look for a good open space that has a good floor, a good wall, and some things in the foreground and in the mid ground, you need like ranges of depth than you're seeing. But something like a tree or this, or a parked car or something like that, or create your own markers that you're gonna replace. But don't go near reflections, okay. Like shiny glass, anything with reflections, wet roads, anything that's going to mess up your track because the camera, the software doesn't know the difference between when a reflection comes across or if the outtakes actually moving. So reflections and shiny things are gonna jack your track up and what Jackie track. Also stay away from things that move a lot. Like if you wanted to track on a tree or something in the trees, like you're trying to put something in a tree and all of the tree leaves are moving around. Or if you're filming and there's a bunch of cars driving across without something that's in the scene statically, it does. The software doesn't know the difference between if the camera's moving and the objects moving. If there's not something else to relate it to, keep you're seen as simple as possible, right? That's, that's the ultimate key there. You want floor and you want a wall, and you want some objects that create depth in your shot. Ok. So that's super simple. So I went outside with my cell phone and Just shot walking very slowly, a shot this row because I knew the roads got some marks on it. It might be good to track. I've got a background. I've got grass, nothing. I've got a sign over here. But nothing is really like moving a lot, you know. So I thought, okay, this is good, nothing shiny. This'll be a good track. And it was, we're going to cover the absolute best practices here. There's always a way to figure stuff out. But if you want to have the least amount of work in your scene, you can fix your scene in the in the planning process before having to troubleshoot later. If you get it all setup in a way that's going to work easiest for you at the beginning, then you're going to be golden, right? So basically, the iPhone, I think, shoots at a normal double shutter speed film rate of 24 frames per second. If you do a higher frame rate, obviously that's going to give you an easier track. Theoretically, because the distance between frame to frame that your point's going to move is going to be less, right? So it's going to have an extra frame inbetween. Each frame that's going like oh, there it is. Oh there it is. Rather than like moved all the way up here, it's going to say, Oh, it mood right here. And instead of all the way, right. So that'll make sense. But then it also doubles your render time and doubles the calculations. And then you also have 60 frame rates. So it's a bit on what you want, really. It's, it's you know, what you're going for. So but 24 hours, what I like to use it mainly because it's half the render time and half the calculations because it has to calculate each frame. So basically you can use fancy cameras and you can use low end cameras, and you can use lights. You can use the sun, basically, which is know what you're looking for and know what works best, right? So natural things in the environment can, can work good, thanks with texture, like a road versus, you know, with super flat white. I don't know, infinite floor kinda look or something. I don't know if you're in a studio or something with the sidewall. That's why you see movies with green screens. They put tape on stuff and they put C stands up with like no pens and markers on it in the foreground. It mask them out later, but they need them there to create that parallax between the two points is if they just tried to film their person walking with his background, it doesn't really understand. The person's moving in the background's not it. Thanks. I don't know what's moving and what's not. The camera's moving so it doesn't track, right, right. But if you have static, make sure you have static things and you've seen that it can create parallax width and that's gonna be the most important.
3. Tracking your footage with C4D R20's Motion Tracker: Okay, first thing we're gonna do is a bumps and for D, So that open rainbow window customization layouts, layouts. We're gonna go to the motion tracker and it's going to change your layout. So now we're going to click motion tracker down here. And it's going to open up this footage tab. And this footage tab is the footage thingy here. Click this, and you're gonna drag your forage in there. So we've got our footage which was right here. It's just straight from my phone. Boom. And it's automatically going to update the frames and also the resolution. And this is important. We need to change the resampling. You can turn to tell it looks kinda puppy. How we need to up our sampling up to a 100%. So what that's gonna do is it's going to make this actually ten ADP and not buffer down to 33% of that is going to slow things down, but it's gonna give you a cleaner track. Ok, so we've brought our footage in fairly simply up to a 100. If you want to edit your scene within here, I don't recommend doing it right here, typing in the frame Start Here. I will do it in outside program first and then bring it in. But if you have to edit what you're doing here, our actually edited down here. So let's say I want to start a second end. I will cut 24. So, but do that first and then I come to here and I do them. Okay, there we go. So now if we go the beginning, it's just going to start right there. Okay. So it's the same as it's the same as cutting it. But now, because I've done the frame starts at 24, it's not going to calculate from 0 to 24, which it's still would I believe, but I'm not sure even if I started as a 24, so but we don't need to do that. So we're going to undo that. Just wanted to kind of show you the good. You can edit your video. Just like that within there. I've gone too far back. And too many times bring off for the gin. And if you need to, if you don't want to go to the custom layout, where like I did, you can actually just bring in a motion tracker right here, this much tracker tab and bring it in right here. But it just isn't pull up all these buttons and stuff for you, which I really like and this tracker view window here, I think it's very helpful. So I like to go to the layout mode. You don't have to. Okay, so now we've got our genome, we're going to track for the auto track. We're not going to manually track. For the manual track, it's pretty easy. You can get off of this select mode. You can just right-click and say create user track here. And it'll make a track. And then you can go into here and you see it's got the track right here. And from that you can click track backwards, forwards and track back from the forwards at the same time. And that'll go ahead and make that. And you can choose a circle rather than a square. It updates here. It doesn't update here. So it just changes, just search radius if you want a circle instead of a square. I'm not sure what scenarios that works best, but I like to just leave it as a square, but we don't need to manually track anything. He just wanted to show you that you could. I could right-click here, like on this point, create user track. And then I can either move it in this window here or I can move it out here. And so what we've just created, this is the motion tracker view. So this little box, everything inside this box from here, everything here, that is what it's going to look for. So it's gonna look for the white to be like this, in the red to be like that. And inside this box is the search radius. So let's say that this is your screen on your camera. And you put a piece of tape over this little square, right? So if you move, oh, you are a, you know, obviously the triangle is now outside of that search radius, but it goes frame to frame. So if you have like a steady ish move and you can hit manual, track your track forward because we're on frame one. It'll track that. So from frame to frame, it's going to look for this exact pattern inside this box, within this box. So as you know, if your object starts out here and it ends up, up here, it's not a big deal as long as from frame to frame, it doesn't move further then from here to outside this box in a frame. If it does, you need to increase your frame, which you can do. Grabbing this little dot here, it's kind of hard to see, but there's a dot there. But now that you've increased, if you increase that too much, then it's gonna say, well, maybe there's something that looks a lot like that over here. So you can say, well, maybe that's it and maybe that's it. You know? So if you don't have a really good objects in your small window, you can really start messing up if you make a search radius too big. So it's really about finding that balance between something that's unique inside this box is something that is as small and upsurge radius that it will catch it from frame to frame, but not so big that it will get confused. And finding something that's not too similar, like if I try to track that, just white, that's going to be tough because it might see that right there and think, oh, there it is, that's it. Same thing as y. This is y here. You know, if I do like OCI and like okay, there we go. But it might say, oh, that's it right there. Right? So you understand with sand, it's gotta be unique. New York, unique enough that it knows it's looking for a specific thing and within a certain search radius, right? So you should look for objects of sharp contrasts. So I see a lot of motion trackers and things like that have a lot of triangles and stuff in them. And black and white. It's gonna go for color and contrast and shape of the pixels basically. So if I'm looking at something that's like that, that looks an awful lot like that. Right? So I think you understand what I'm saying, but I believe that this is going to be pretty good track. So we'll hit track forward here. Okay, it's done already. Hit play. You can watch it. Amazing, right? Fantastic track, route the bat, easy-peasy, no problems. Then, then it's all screen and it's killed, it's dead. So that's fun. I just want to show you how that worked. We are not going to manually track, but if you do need to manually track something, you can you need probably at least if you're only going to manually track and not automatically track, you need at least, I'd say at least eight good solid track points so that you can get away with six. I think you technically have to have four. But I would recommend at least, you know, eight to ten good tracks. And because that way you, because what you're going to need from a manual track is you're going to need to know the depth z here you're gonna need two points to show this is the z-axis. You're gonna need two points to show this is the x axis. So point here, point here. You're gonna need two points to show this is the y axis. So point from here to here, there'll be a wire here to here, wherever or backyard here, you know, that kind of thing. Where you're going to need to show z, x, and y for your Solve. And you're gonna need good points. That organ to have good parallax and lead to calculate Well, so I guess I'll tracks, I like the auto tracker because it's really good as powerful and less manual work, which is always good in my opinion. Not always, but it does work really well for this. So what we're gonna do for this is we're going to use the auto tracker. And I'd like to say a 1000. I think it would be good. Minimum spacing, I'm going to say 19 is fine. That's the default. I do not know what the units are. 19 what, i don't know pixels, but perhaps I don't know, but basically it's the spacing between points. So let's all right, I'll go ahead and I'll click Create Otto tracks and I have other replaced last tracks on. I replace lost tracks means if it gets to the point where that track point over here when offscreen. Now that track is dead, it's gonna create a new track points somewhere and start tracking that so that there's always a 1000. And that way you can have over 1000 tracks when it's done. But if the track is not out or replaced, it's just gonna kill that and it's not going to replace it. So you'll only have 1000, which can be good and useful too, but it's not bad to have more tracks really. So I'm gonna leave auto replace lost tracks on because I know some things you're gonna go out screen, some things you're gonna come on screen. And it's good to have track points for that stuff sometimes. So I like to do auto replace lost track and it's hard to say. But what I'm gonna do to show you this minimum spacing is I'm say create auto track. Boom, hit that button. We've got track points all over the place. You can see they're not grouped up together, right? They're this far apart, which is probably 19. I don't know what that is. But basically, if I had made this five and hit auto track. Let's delete honor tracks, create other tracks. There you go. So now you can see they're a lot group together a lot tighter. So any part that the, there can be like right next to each other. So create all these track points really close to each other. So if I did like 50, delete other tracks, create our tracks. Now they're only going to be spaced out 50 apart from each other. So you can think about it like at 1, draw a circle around it. The next point that's going to be tracked is gotta be at least 50 away from that, right? 19 is the default. I don't know why, but it seems to work well. And I think that looks pretty good to me. So what we're gonna do now is we're going to click and this is normally grayed out, but because we went into manual tracking, it's not. But pretend this is grayed out because it will be for you if you just go straight into auto tracking. We are going to with a thousand we've clicked here. We wanna go to the middle of our scene. Because I like to go to the middle because the way I shot it, I walked slowly up the road and to the right a little bit, but I kinda kept this is kind of the center of the screen a little bit, right? So what I want to make sure I do is I go to the middle here, about 40, we'll say. And this is where I want to create the points. Don't create points. Because if I create points in beginning and track forward and always going to be points in the beginning that aren't at the end. And if I go to the end and create points and track backwards, and I'm just going to be points at the end that aren't at the beginning at all. But if we go to the middle and create points there, I know that those points I created are probably going to be there for most of the way until the beginning and towards the end. So I know that these points are going to be onscreen longer than if I started the very beginning, in the very end. And anytime you can get points to last longer across your whole scene, it's fantastic. So if you can make sure you keep stuff in the shot the whole time, that's going to really help your track select this little spot. I kept that on there pretty close but didn't get it the whole time. But there's some spots here that really sitting there the whole time and that's going to really help out. So we've got this. So now what we're going to do is we're going to leave these are normally grayed out. So if we're going to ignore those and we're just gonna say audio track forward and backwards because we're in the middle blue. So that's gonna start processing those frames left and right, forward and back. And it's gonna go framed, reframe with each one of these points and calculate that. So get a cup of coffee or open up some silly videos on YouTube or something. And so it'll take a while, so I'll come right back as soon as it's done. Okay, so one thing I want to show you before do this track is mask, right? So let's say I had like, this is the sidewalk and this is a skyscraper here and there's like an office building and there's windows here in their reflective and I don't want that to track. There's people over here and they're walking and moving around and they're gonna mess on my track. So I want to mask them out. I just want to show how to create a mask real quick. It's right here. So you're gonna click this create Mask button right here. And by default, it is set to exclude. So that means we're going to want to draw a box around things we don't want in the scene. So let's say I didn't want any of these trees to be tracked because they're like moving or something, right? And I'm like, okay, let's just check this out. Click, click, click. And ego. And so we've got this box here, and we got it as you get selected over here, we can click this. You can see, you can say exclude. So now you see there's no track points in there. Or you can say include and it's only going to put track points in there. So this can be useful. And the way it works is actually auto updates. So if you scroll down you scene and you just want to move it around to where you had it, right? It is actually going to automatically animate that for you. I don't think that was a good example. So let's say I wanted to bring this up. And here I want to take this down. Cell C. Go up here, keep this out of it. You know, let's say I want to move it and you can tell it, I'm actually moves in the keyframes and place to place. So that's pretty awesome, but we don't need a mask for our scene. I just wanted to show that you could do that if you needed to mass something out like somebody that had reflections or something, you know, is gonna give you a bad track and bad problems and you don't want to add ope track there. Go ahead, mask it out, move it around. I'm actually keyframes it for you is really good. So I just want to show you that before we go here and we say, I don't trick. I really want more tracks here. I'm actually just going to up this D like 2 thousand and say create tracks. There you go. That gives me a few more. Okay? So I'm going to click out of trek. Okay, that finished. Let's see how this looks. Subtract points here is somebody going in and out. That's okay. It looks pretty good. Nothing's really sliding around too much. It looks like we're mainly just losing tracks, but nothing's jumping around or sliding too much for just repopulating and losing. And that's OK. That is a pretty solid track. So Yeah, I dig it. I dig it. So what we can do to see your track health kinda thing is clicked this motion track graph view. Or it can go up here most checker, most checker graph view. And then we're going to hit this to frame all of them and this is all of our trek points. Okay. So I don't see any that survived the whole time though. Let's turn this up to 200. Okay, so now it's only showing one set survived the entire time pretty much. Let's look. Okay. So these are my tracks that alone went from frame 0 all the way to the end. And that is great. We've got some good green ones in there. We're gonna wanna do is just kinda clean up some of these red ones. Where you can do is delete them. Or if you think you can fix them. Let's see this one. That's, that's an, a bad spot. We don't want that. But let's say there's one that you needed to fix and adjust. Let's see if we can find one here. Let's go back down to 15 kilo. So we've got this one here on this point, right? So let's look at the graph you for that. And let's isolate this one. Right here. It goes to frame 125. Zoom out as baboons and never understood the controls in this slide over here. Ultimate clip middle mouse button. So you can see right here, this ended at frame one, 25, right? It's okay. It's gone. 27. So here it is. Frame 128. It goes away. It's off. So what we're going to say is we're gonna tell it that actually. So b right there, there's things where it was, right. Okay. It's more than middle. The I get it right where you want it as wrong editor there. And it automatically updates and look, it actually tracked it. I think to lean loss it again, let's say quit. Insulated. Okay, here's our trek. And we want right there. And so, okay, so it lost it, right? Let's leave this because my outer tribe, I made a stupid, so lost it and we know it wants to be right there, right? So we're gonna put that in and you can see I'm actually updates. And that's really cool. So now it play it again. It will go back and fix it. But like, I don't know why that's there. That plus the source one. And ECP more here. Okay, now we go to the middle. This is kinda freakin out for some reason. We want to auto update, depleted itself. Play. Okay, so there look pretty solid track. Oh, it's got enough. We're gonna go put it back where it belongs. Let it update, play. And there we go. Now I had attracted ended at a 127 frames and now it goes all the way to pretty much the end. And that is fantastic. That's what you wanna do, that's how you fix a bad track. So now we've done that. What I'd like to do is save, save incremental, just appended number on the innovative in that way in case you need to go back to a former version. You can. But now we can go to 3D solve. We wanted to a full 3D reconstruction. We don't wanna mess the nodal pain or a planner track right now because the camera is actually moving is I'm going to try pies we're going to use for 3D. Reconstruction and that's what I was talking about with parallax and everything. That's what we want. For the focal length. If you know your camera's focal length, you can definitely put that in. If it's constant, it'll say constant. But if you zoom in, you definitely wanna make sure you have unknown clicked, but not constant. Okay? But for this one, we know that the focal length is constant because as my iPhone and I didn't zoom or anything. So we're gonna say unknown but constant. And we're gonna say lock solved data. And we're going to click Run 3D solver. And it's gonna say it's running 3D solver. And we're going to hope it doesn't crash. And we're gonna come back to that when it is done. Okay, there we go. So now you can see we've got all these little dots all over are seen here and this is perfect. Was going to go ahead and hit play. I'm going to put this and watched these dots ignore this grid right now. But these dots are going to indicate track points and how will they tracked until nothing is sliding around or anything, we've got a really solid track. So this is really awesome. This is really good. And so you've got the bright green is your best track. And then the size of them indicates their distance from the cameras. So the bigger circles are closer to the camera. Tango circles are far away. And so read the brightness from green to red also indicates distance a little bit, but more so, just track quality. So we've got this, you can see all the way through, this is fantastic. We change our camera view. You can actually see sit on and play it. But you can see we've actually got light from the top down. We've got a camera here and we've got you can actually see the edges of the road here is really wild. And you can tell like over here it's the grass and the sine is this tree over here, and this is this tree right here. Back here is the, this little area right there. In saying that, well, all right, they're really, really cool. So we've got this. And from the side you can tell it and thinks it's all sloping uphill a lot more than it really is. Same that this front angle here. So one thing I really like to do is I go in here and you can see all of our auto paths and it sorts them in order from very best quality to the worst quality. But you see how it gets brighter red you, okay? So what we've got going on, what we can do now is go to our 2D track. You need to do that. Let's go to, we're going to go to the reconstruction now this is really cool and this is what I like to do. I have to go to preset preview. And this is your settings here. And I'm gonna use a generate mesh. So that's gonna do, is that is going to actually create a mesh from your track points. And I'd like to do preview because it's going to be the fastest one to calculate it. It's not going to be very accurate as far as like reconstructing these trees or anything, we don't have enough data there, but we do have a lot of data on the road. So if I can get it to create the road for me, that's gonna give me a great jumping off point to know where to put my plane if I wanted to be the ground. So here we go, look at that. We've got our measure. Play. Boom, that is awesome. That's not sliding. It's not moving at all. I mean, that is perfect. So now we have a good base of where our road is, right? So this is what, if we wanted to put something on this road here, we could definitely do that. Fantastic. Okay, so what I like to do next is I go to motion tracker here. Let's make sure we're back in our layout. Motion tracker. Okay? And what I can do is create a vector constraint or a planar constraint. Okay? And so a position constraint you can also create as well. But that's going to be for more if you're filming around something like if you had like a statue or something, you rotate it around. And you know the position constraint you will pick. The University did an example with the statue and they needed the position constraint right on the nose because that was kind of the focal point at the scene throughout the most of it so new that your camera was orienting around this point, right? So that was good. Basically what the constraints are going to do is it is going to determine the relationship between your track points and your camera, like rather than since right now if you look at your camera, it is at 000, is perfectly at 90 degrees is 000, and that is not how we really want this scene to look. Basically, when you add constraints, you're switching it to as if the world is the world and the cameras. The camera right now. The camera is static and the world is moving around it, right? But we want the opposite. We want the camera to move around the world like it isn't real-life. So with the vector constraint where we can do if we click our emotion points here, but we can choose good points. So what you're going to look for with a, I don't want that. It had this selected. I want a vector constraints. So what this is gonna do is let you pick your axes. You can do x axes. So you could go from here to here. That would be your x, your y, which would be your height. So you can go from here to here, some embezzling up like boom, boom, like this point to that point would be a good one. Z-axis. So something from here to there, your floor. So you're gonna tell it that this is the floor, this is the riot, and this is the up and down. And then you can also do the distance that way if you needed to create a new scale. So we're gonna go boom, boom. And that's gonna be my z, right? Because that's the floor, I definitely get that. So over here, we're going to tell that that is Z and we don't know the length of it, so we'll leave it unknown for right now. Okay. But if we did, let's say from there there say that it's about, I don't know, five feet or something will say known. And now we can type in the distance. Let's say 500. Ok, just for fun. Which is a lot more than what I said. But that's fine. So now we have a distance there. And so now if I wanted to go from here to here and say that was my x axis. And that, that distance was four times that we could do that. And then we'll say from here to here is my y. And it is, we'll say unknown. Well actually we're going to go back to this one and say unknown on all of these, we don't really, we're not doing anything to scale, but if you had something like we used to do a lot of stuff with them, drone footage or we do RVs. We'd fly over an area and we would put in the model of the building before it's built. It would need to be to scale because that's what they wanted to show. So we would have to measure out the space. So if you wanted to do something to scale, you need to actually know these measurements. You can't just guess it because that's gonna mess stuff up. But if you get them accurate, it's gonna help out a lot. So that's a pro tip. If you want to measure out your scene, that's going to really help you out and use these vectors to create that. But right now let's do it without it. And let's just take a look at what we've got now. So now if we hit play, you can see are all of our points are exactly the same. If you tell our grid has moved a little bit. So let's take a look at this outside of the area. So now we see if you look at this from the front view, our road is flat, which is pretty good. That's what we wanted. This is the z depth. And then you can tell here it slightly ramps up like this does, but it's pretty flat. And the cameras actually pointed down a little bit, which is what was really happening in real life. So now we've got this where the camera is moving around, it's not stuck at 0. All the points are stuck. And now the camera moves, which is what we want because that's really how it wasn't real-life. So now, if we add in, say a cube is going to be gigantic as I didn't set my scale. Let's bring this down like ten. That was painful to watch me do that, wasn't it. We are going to make sure when you're everything else clicked, we're going to select our object. Zoom out here. And we're going to drag this in Raqqa and we're gonna plop it down. You can see we can put that right here on this road. Nice and flush. So let's say we put it towards the camera right here. Let's zoom in. Let's turn all these points off because they kind of get in the way. Hm, okay, so that looks pretty good. So we have a nice flat plane that's really solid, is rotated. Just a genius, but if you turn these on, it kinda big. So what I'd like to do is like to go into the motion tracker and I go to duty solve, display, circle, radius, a particular data 0.5. And that's going to bring all of these points down. And sometimes that just makes it easier to see patterns. I get it's going flush or anything like that. So that's looking pretty good. So now if we say we hire mesh and we just have this cube sitting here, it should look like this cube is. Perfectly sitting in space without wiggling around anything. That's looking really good. So now the cool part is, now we grab this cube. If we move it in Z space, like closer to the camera, it's going to move closer to the camera. Backspace is gonna go further away. You go up, it's gonna go up, right? Is gonna go right? And so it's really obvious when you stretch things out. So that's cool. So now we've got this big platform that's going down this road. You till the road bends a little bit. So let's do that and go down the road. Up the road and it's yeah, that follows the horizon line perfectly. That is fantastic. That's what we want. So by at play, This is going to just sit there. And there's a good solid track. And because we had that the vectors constraints and because we had the the good reconstruction plane, just to double-check, we can see that we have this really nice track and it's really easy to bring an object in, get it lined up oriented, and just have a nice base to start off with. So that being said, let's get rid of that cube and let's add a plane. Now there's another way we could do this if we wanted to do it instead of a vector, which will get rid of that. And it's gonna undo everything we just did, which I thought was a really good result. So I'm not super happy about getting rid of it, but we're gonna get rid of it. And what we're gonna do is actually use the planar constraint. So if I use the planar constraint, I can go boom, boom, boom. And I've selected three points that are all on the same plane, right? They're all going to be the floor plan. And that's gonna be called y. And that is what we want to determine is the floor. So now I've got this plane here. You can see this triangle that I've built. And it has already done what the vector constraint was doing, which is lock everything down and have the camera move around. But where you can tell the difference between the, the planar constraint is it there's still kinda got this angle ramp up a little bit which I didn't like as much as I liked. How accurate and flat the vector constraint in. But one thing that the planar constraint has, the, the vector constraint did not have is this little ability right here. So let's click this. Under selection, you can see the track points you have. Make sure they're all green, that's what you want. You can add more. But one thing you can do is you can say create plane. And that's going to instantly make a plane. Let's hide our scene measure. Instantly make a plane exactly where you want it. And make sure you click out of that. Go you plan, you can rotate that. And you can just go ahead and move that around. There's going to be oriented correctly. It's just that out. Stretch it out this way. There we go. That's better. So there we go. So you could tell it looks like it gets a little smaller. That's because if you notice here, especially in this view, you tell it when we move it, it's actually going below where the ground is. So that's why it's looking like it's getting smaller as it gets back there. And that's not good. We don't want that. So what we can do is either add more constraints to it. Let's go to our low constraint thing. Add another one. Let's do one way back here. Add that guy. Now let's see. Now this plane reacts. Click out of that to go back to this yellow. Get out of here, click out of that guy. Say, leave that plane. And you need to create another plane from that. So now we've got a plane that stretches all the way back there. Make sure you click out of this is really difficult to click out of that. To remember you have to do that. But there we go, we've clicked out of that. And now you can tell we kinda have this plan is kinda scaled more normally, right? So it's going all the way back there. So there's shouldn't look like it's getting any smaller ups. The mi e. So as it goes back, it doesn't look like it's getting smaller anymore. Right. And saying the same width as the road up in the front is in the back. So that's good. That means r or plan. If you look here, you can tell it's kind of going along the ramp of the road even better, right? It's not going down underneath here anymore. So there we go. We've got this nice big plane right here we can do and which can stretch that out for now, we're going to hide arsine mesh. It's good that it's still there. And that's fantastic. So now we've got this nice plane here, 011. So now let's add a cube loop. Shrink to keep down notes. Hit tab to go between those faster. So now what we do, now that we have this plane, we can go to the, back to our out of our motion tracker view because we're done with those options that we need from that. And with your plane, we can go to our rotation here. Sorry, let me go to a plane and go to our coordinates. You can right-click this r, say copy, go to this cube rapidly are say, paste. And now our queue will be rotated and oriented exactly the same as the plane. So if we bring this down, let me zoom in here. You can see we can get it perfectly flush along that plane. So it had a little bit of rotation with it, without it, it wouldn't be perfectly flush. It would be. Exactly even in parallel with the 0 floor. So now we have this good plane that we can make sure everything lines up with perfectly so sick r cube. It's good. The object tablets at little 2D and do like a one and not even that as 0.5. And this just do three segments. That's fun. Okay, so we've got this cube now and move this cube in world space. It WAS flop to that for backwards. So I decided here we go. So we've got it, let's pull it up here. Let's grab this plane. Let's pull that up just a little bit like that. Okay, we've got this cube. So now I'm going to show you exactly how to do the compositing. So this is the big thing I'm gonna first I'm gonna show you how to do it in redshift. Then I'm going to show you how to do it in physical renderer. So compositing with redshift is really stupid, easy, it's so awesome. So here we go. Let's go to our theme here and make sure we are said in redshift chick that, check that off. We are going to right-click our plane. We're gonna go to redshift tags, redshift object. We're gonna go to Matt. Check, general enable, check, shadow, check. Yep, yep, yep, this is all good. Apply secondary rays is affected by MATLAB, include puzzle Matt's, we don't need to worry about that right now. Secondary rays we can turn on, that's going to be things like global illumination and things like that. I believe I like to have that on MATLAB and puzzle maps we don't need right now. But, so what we can do now is go to redshift lights, dome light. And so what we need to do for our dome light is either if you have a 360 camera, you can go your scene. If you know where you're gonna put your object, plan that out. Take a 360 image of your environment and boom, you've got an instant super accurate HDR. But I don't have that. I shot this with my phone and I didn't even download an app or you can stitch those together. Those exist. You can definitely do that. But honestly, I said, you know what, I've got a street, i've got some trees and I've got a sky. And I know I've seen HER ours that have those things in them. So let's try to find one. It's very similar. And so here's what we're gonna do in this is only for the reflections, right? So if you see what I have here, I found this from HER Haven, which is a free HER website. It is something that has kinda an overcast day. It's got some trees and it's got Road. And so that's really good. We don't need to copy it to the project location. We can see here over casts some green and some row that's about the same color as my road. It's really good. My objects aren't going to be super reflected, so it doesn't need to be perfect. But if you already do something like a cool car and automotive reflective metallic material or something like that. You're going to definitely want to focus on getting a good HDR for those accurate reflections that's gonna really help sell your scene. But if you can avoid your object being reflected, that is also going to help a lot because that is the one thing that gives it away more than anything for me, for my opinion. Kinda Uncanny Valley like like I know that's not right, that there's something about that doesn't look right. It's almost always the reflections. So now we've got that. So if we render right now, you're gonna see that it should just render it on black because we have a light in the scene. So it's going to let r cube will let this boot up. So you see right now, we've got our cube and it's sitting in a dark somewhere, which is just our HDR map. So this isn't right, this doesn't look like our scene. So here's what's going on. There's two things you can do right now. One, you can go back to that real quick. Let's go back. And you can check out that there is in fact an alpha channel, which there is not. One thing we definitely need is an alpha channel. Actually Hold on, let's not do this yet. Okay, so you can see that this is not correct. So what we need to do is we need to go into our dome light, scroll down to back plate, enable that. And this is where we're going to need to take our MP4 footage or whatever it is or video footage, take it into Premier or encoder, and export it out as a J peg sequence. I'm gonna show you how to do that right now. Okay, now that we've got that, export it out as JPEG sequence, we're going to just find that exactly where we want it. Let's see Baba, Baba, JPEG sequence. Click the first one, open, very important. Do not put this in the search path. It will not take the entire sequence. It will only take the first one. So hit no. Now we're gonna go over to animation and we're gonna go mode simple. And we're gonna say detect frames. And there we go, start frame 0 and frame 264, which is perfect because that's exactly how long your video is. So now, if I've got a redshift reinterview, hit Render. But we have a bomb. There it is. It's in our scene over top of our footage. Now you can tell the lighting is not right. But here's what we're gonna do. One thing we're gonna do is make sure that we turn on alpha channel replace. And what that's gonna do is now when we render this, because if you notice this is a little brighter, more saturated than this, right? So here's the thing. With this on. We now have an alpha channel and you can tell it's showing this mat. And that's not what we want because we want this met to only catch shadows and we only want our object, our cube, to show up. So underneath the redshift tag, underneath this map, we're going to say effects alpha affects alpha. There we go. Now our plane is gone. Only r cube and r shadow is there. So now what we can do is we can take this footage. When we render it out. We'll save it as both an image and an alpha channel. So now if there's something we weren't the color corrector change in our original footage. We can do that and then just overlay this image, just the cube and the Shadows along with it over top of it. Instantly with the die, everything else will just be invisible and we'll just plot Brian in there and you'll look right. So now we can adjust it, the lighting and stuff for arsine. So let's go ahead and just take our DOM line. And let's just go up the exposure just a bit. Maybe about like that. Okay, let's create a material real quick. Redshift material, material. Throw that on the cube. There we go. And if you look our kids floating just a bit. So actually wanted to bring that down. Just a bit. So it sits on the ground there. Yep. And you see the reflections don't match up. This this doesn't look right because they're not correct because my HER is not the same, but there are similar but this is a super glossy right off the bat. So we're gonna go in here and it's shade a graph. We're gonna choose paper and then we're gonna take the back translucency, backlighting translucency out because we don't want a light to shine through it as if it was paper. So I've got this bright white objects. So now we realize that, oh, maybe our exposure was fine. Double-click the zeros, that will right-click these arrows, bring it back down. Okay, there we go. So now we've got our cube here in our scene. The lighting is the same direction really, but honestly, we really want this to kind of be in the shadow as well because all this isn't a shadow wouldn't make sense since all of this isn't a shadow as well, this shouldn't be so bright. So what we're gonna do is we're going to adjust. We've honestly could just get away. We either put a
4. Composite your footage with After Effects: Opened up After Effects, and now I'm going to show you how to import your footage and composite it into you seeing with your 3D stuff in there so that you can color correct both of them simultaneously or individually, and also add some motion blurred here object just to kinda help blend it into your scene rather than using the motion blur to redshift or to physical renderer. So I'm gonna use the redshift version of what I downloaded, what I exported. And I'm also going to use a version that I did earlier rather than waiting for it to finish. This is something I actually recorded this whole tutorial all the way through and realize halfway through my microphone stop working. So that was awesome. That get done with it and find out that I needed to do it again. But here we are with powering through, we're going to do it is, we're doing it again. Take two, here we go. So we are going to right-click in here. We're gonna go to import file. We're gonna grab our footage, import that in there. We're going to take this and we're going to drag and drop it over this little icon right here. And that's going to make a composition from that footage. So all the frame rates, everything will be correct. We're also going to now import the, export it out jpegs, right? So we've got some I want, so we're gonna import our export it out jpegs. Here we go. So you just need to grab one of them in the sequence. You don't have to have the very first one, but make sure that PNG sequence is selected. So we're gonna import that. And you see it says frame rates 30, that's fine. We're gonna grab that and pull it in over top of our footage and whammy, there we go. So now we can actually turn off the background and you can see there's nothing here except our object here. And so if we hit this little button that's gonna allow us to see the transparency that shows that this object has an alpha channels. So when we turn our footage back on, there it is. So now Space-bar, we're gonna preview that. And I've got it set over here in the preview does skip every two, every two frames. So just so you can play back a little faster. But you can see here we go, there's our footage. It's compton, the scene. And so now if we wanted to color correct arsine, we could do two things. One, we could color correct them simultaneously, together by adding in adjustment layer right click down here in this space. And adjustment layer, go to effects, go down to color correction, go to luminary color. We could add a lot to our scene if you wanted to. Something like that. And they're gonna match now, right? So that looks pretty cool that they match. This is a silly let, but we'll do it for now. But if we didn't wanna do that and we just wanted to do them individually, we could take our object here and let's say I decided now that I look at it again, I don't think my object is dark enough for my shadows are dark and f for this. So I'm gonna go to limit your colors. I'm gonna go to colour wheels. And we've got the shadows. I'm gonna pull them down. And that's just gonna make my shadows. Darker on my object here. And I can do my highlights could be higher or lower and I think higher looks. And it looks okay right there. And move my shell is just a little bit on the blue side because that's what real shadows are like. I can also up the contrast here. But we're going to actually just gonna leave that 0. And I'm gonna bring my shells down here as well, which is just going to double dark when I seen here. And then the contrasts Backup. Yeah. Okay. So now I've got that in there and I'm going to turn my back on to blend them both together. This looks pretty fake because there's no materials or anything and not sure that this is really unnecessary, it looks better without that. So I'm not gonna do that. I'm just gonna leave it like that. It's it's not, it's, you know, whatever. Okay, so now we've got these colors together with this funny, fun, let on them. One thing we wanna do is we grab our footage here. And I like to add just a little bit of noise to it. So one thing that we can do if you have edu Kramer's, I'll actually, I'll wait. So we're gonna add noise to it and I'd like to do like 2%. And that's just going to add a teeny bit of noise to our object. We worked so hard to get it to render clean. And then we're going to add noise into it just, just a bit because you look at this, this has noise. So it blends and a little bit, we can tell this is still a little too sharp. And then go a little further along where the camera is moving a little more. And we'll look at this. And we can say that looks pretty good. So we want to go to, we're going to hit control space. And if you have Fx console, that's going to bring this up. If you don't have Fx console go to video copilot.net, downloaded. It makes finding Plug-in. He makes finding effects so much faster and easier. It's awesome. But so in case you don't have that, we're gonna go over here to this effects and presets button. We're going to click that and it's going to lag out because it always does. It's gonna load all of our effects in there. And then the reason why I gained your Kramer's plug-in better. Be sure to take that out. Take that out. Not too far to that. Okay, so now we've got this opened up, so we're gonna type in motion. Okay, so there's two things we can do. We can add CC forced motion blur and pixel motion blur. I prefer pixel motion blur. I think it gives a little more accurate results and I've had less weird glitches with it. So we're going to take that and we're going to drag and drop that onto our footage of just our box here. Okay? And so now we have this. We want the shutter angle to be 180, and we want the shutter samples to be 16. Vector detail of 20 is fun. You could tell that's going to slow things down. Cc forced motion blur is faster, but pixel motion blur is pretty cool. Basically what I'm doing here is arsine kinda has some blur to it. It's not a whole lot, but a little bit. And so, but our object had 0 blurred all one button are both. Our object hit 0 blur whatsoever. So I will kinda wanna match the motion blur or seen. I really want to try to find a clip. Or the camera moved a lot. There's not a whole lot going on, so it's not going to be very drastic, but it's more obvious and faster scenes. But basically the way pixel motion blur is going to work is it's going to look for where the pixels move so far, it's gonna like try to calculate and guess what the pixels will look like, blurred together on the edges. So let's say we want to open this up and do 360 shutter, which should mean shatter angle is completely open. So you're gonna get the blur is that it can get basically a shutter is normally set to a 180 degrees, which would be if you're filming at 24 frames per second, your shutter would be one over 48 or equal to a 180 degrees and 360 degrees. You shutters all the way open. 0 users all of your clothes. So higher the shutter speed, the bigger the shutter angle. Now, sorry, higher the shutter speed, the lower the shutter angle. I apologize. So we're not seeing a whole lot of difference here. But you could tell there, you see that. So let's open this backup 360. You see this little blur that is added here. So we turn this off and that's just really sharp, right? And something about being really sharp and the rescue scene is not sharp really draws attention to the fact this is not belong. But you add that back in and now it's getting a little blur with it. They kind of just helps sell it that is, in this scene. Let's bring this down, see if that helps. I can't tell the difference that 2640 samples. Let's see difference or snapshot off. So now if we look at that snapshot, you can tell is see the edge, the difference there is, it's got a little blurred to it, which is going to help sell. It's moving in there. So there we go. We're gonna leave that on for sure. And 360. And what we need to do now is we can literally just render that out. Or we could go in control things. We can do more effects and things we wanted to. But there you go. That is how you. Composite and it's exactly the same with the physical render thing. You literally once you have an export it out with an output channel, the way we did it, you literally just plop it in right over top of your footage. So you got your footage in there. And then you add your scene, your 3D object in there. And it's just gonna go right over top of it. No problem. Easy peasy. One thing I wanted to be sure to show is how to export this when you're ready. So if you have your comp open, you're going to hit Control M. And that's going to bring your render queue up. And it's going to have your little icon down here and a little checkbox. And we're going to want to do if you're going to want this to be a H.264 is Q and AME, which is Adobe Media Encoder. But from here you could select this and do certain things like APIs or PNGs sequence, stuff like that. If you wanted to. I like to do stuff for H.264 because it's going to end up on YouTube and Vimeo. So the smallest file size when you hit q in AME with it, checked here. That's going to open up encoder and it's going to already populated with something like this. So now I have my comp here. I can click this icon. And that's going to open up a dynamic link between the two. So the cool thing about using encoder is you can render it out without locking up your programs. So if I get this going, I hit render on this. I can go back into After Effects. I could even keep messing with this project. And they won't live update or anything. But it'll, I keep messing with this project and working on it and we're work on the next part or open up a whole another project. But if I try to render it out inside of aftereffects is going to lock up the program so I can't do anything else while it's written during ice. Gotta sit there and wait for it to get done, which is fine. It has its place like at the end of the day or whatever. But I really like encoder because you can do that with both premiere and after effects both. You can send stuff to it and let it batch, render all the ones while you keep doing your thing. Okay, now that that's popped up, what you want I'm gonna do is click this little format arrow down here, twirl that baby down and we're going to see all these different codecs and whatnot, but we're just going to look for H.264. That's what we want. We've got things like PNG, QuickTime, which is just MOV, AVI, which is going to be gigantic file size. And so as you know, we were going to H.264. And for the preset, we're going to just make our own. And so we're just gonna leave it how it is. But we're gonna say, okay, files sizes, right? Square pixel is good. All this stuff, it's matching our framework here. But now we get down to bit rates settings. And this is where we're gonna wanna change things. What you can do for this is, so a constant bit rate is going to lock down that they read. Every single frame is going to be exactly the same quality. Every pixel is gonna get the same attention, variable, bit rate. It's going to change depending on the scene is going to work as hard as he thinks he needs to work in that kind of thing. The reason you use variable bit rate is something like if you have the best example is like gaming footage, like so you have a third person game like Dark Souls or something and you're watching, it's got a whole lot of particles and elements and movement and everything going on all at once on your screen, a lot of movement. A variable rates going to make that look smoother and it's going to the motion blur and everything is just going to look a little better and a little smoother than the CBR. Cbr is constant, V is variable. So one pass means it's gonna go through it once to pass obviously means it's gonna go through it twice. It's just going to be a little cleaner the second time around. But I'm just gonna go CBR. I'd like to do that. And I like to say 2020 is a good broadcasts kinda level quality. It's going to look good on YouTube. No one's going to know. If you put it up to 50, where ten ADP or whatever, no one's gonna tell. Now if you start doing stuff and 4K, you might need to get up into the thirties and forties. You file size is going to be bigger depending on your bit rate by equality is also going to be better the higher the bit rate. But 20 is going to be good for, for 1080 P or 720 P. Wherever you're gonna do. I like to click use maximum render quality. I don't know why you wouldn't unless you're just doing a draft. Which if you go, so now we've got that and then click this output name. And that's going to be where you save it to tracking, compositing. It's all over. Okay? Okay. And then now that once that lows backup and things blew, you get this little green play button start cue. And so there you go. Now that's rendering. So now I can go back and after-effects and do stuff around and it's not locked up, but it's still rendering everything. So that's really cool. That was a very, very big tutorial. I had to do it more than once. But after a lots of testing and trial and error, I feel like hopefully I got you something that you really feel like you learn something from. You can take what I taught you today and actually go apply it to your footage. Here's one good example of I didn't render out this this with motion blur. But I can take cc motion blur, pixel motion blur, plop it on and say 32 samples. You can tell already right off the bat. That looks pretty good, pretty accurate motion blur. So you can see the samples are going to smooth out here. There we go. So there we go now without width. And we can watch that back. And instead of. But yes, we hopefully you can take something from that. And it's going to slow down the render time. I really wanted to make sure that I made it tutorial that a beginner without all the equipment and all the plugins and everything could do and follow along with, I feel like After Effects, it is common enough. They let people have it. If you don't, don't fret, you can use Fusion or nuke for a free version of that, I believe. Just look that up and then the same thing will apply for. You could even use Photoshop, but if you have volition LP, probably other stuff. But if if you need to use something else, composite your footage, you learned how to export it with an alpha channel, which is the most important part. So now you just need a program where you can layer video layers together and they will recognize alpha. So it'll automatically put your footage, you 3D object over top of your footage. And that's really the key, no matter what program you use it for. It should be able to work. So that's good. Hopefully the rant I had at the beginning of the tutorial about all the stuff you need to look out for when you're filming isn't too overwhelming. Honestly. You're never going to well, I'll say never, but normally when you go out and shoot something, everything is not going to go perfect. You're going to have to troubleshoot. That's part of the fun of being in video production. But also know that the troubleshooting is okay. It can, it can spur new information. I did a lot of stuff that failed before I had some success. Figuring out the vectors I think, or my favorite way. The vector constraints are my favorite way to create the axes rather than the plane constraint. Just know what you want to shoot before you go shoot it. You don't have an idea in your head is said to be fleshed out completely, but have an idea. Look for good scenes that don't have reflections or mirrors or water particles and things like that. If you can avoid it. Just for practice. I mean, like obviously, you know, start figuring out how to do mirrors and things, how to, how to go beyond those, the problems. So I'm just trying to show you how to do beginner stuff like stuff that is going to help you go shoot something and track something for yourself, right? And then from there you can get to more advanced things like masking out people in greens, Greens and moving things around and stuff like that. And the stuff in the water and things like that. But definitely just if you created something with this information or anything, or you just want to share a good advice or feedback. Please post in the comments below or hit me up on Twitch, stream live on Twitch every now and then. I would love to do more. Definitely hit me up on Facebook, Instagram, Twitter. Probably Instagram and Facebook are your best, best ticket to get in touch with me. You can always DME and Instagram, it's affected Tron. Same with Facebook. You can look me up. I would love to see what you guys do if you guys you've used any tracking stuff. I just would love to see that. Hopefully I helped you guys out and that's really all I want. I wanted to present something that was flushed out. It wasn't like here's a small piece of the puzzle. I wanted to present every piece of the puzzle so you could actually start putting it together yourself. Go out there right now with him, a 40, shoot something with your phone if you needed to track it, render it, put it together. Boom. I wanted you to be able to do that without having to have like, oh, you need to have a gamble and unsteady rigor and something like this or that. So I had some fun using the camera tracker to do a little Mad-Eye Moody kind of effect here. And I actually use the tracker to reconstruct my face just from track points in 20 points on my face or anything. It literally just filmed around my face a little bit. And then it reconstructed it enough and I could plug it into a volume measure and get some nice shadows and things, but it really just gave me a good reference point to put some objects. This is Brill rough render out, but you can tell there's some mistakes, but that's pretty wild. And that's insane. That's really, really cool. So there's no track points or anything going on here. And I literally just plop that on there. And that's really solid, very cool stuff. And as wild, the points from the track point actually just look to the points from the perspective viewport. And you could see the shape of my face and my profile and everything in dots is just bizarre that it could just actually accurately recreated my face who's kinda freaky. Again, thank you so much for watching. It means a lot. I'm exhausted. That was a huge tutorial. I really wanted to create some and they went from pre concept, shooting it, tracking it, compositing it all the way through for anybody at any level. Beginner, novice, advanced forever as long as you understand the basics and things to look out for, hopefully it'll allow you to troubleshoot and go create some of your own stuff. If you do create something, please share MY tagged me in it. I would love to see the stuff you guys create and I love to create a community here of just advice and friendship in and stuff like that. I'd love to hear from you guys. I love your feedback and everything and just have a great day or night. Thank you so much for watching. See you later. So and if there's anything you see that you think you could do better or you have any ideas, they might work better, please let me know like I would love to create a community of advice here and things that are best practices, you know. So there's, that's the great thing about 3D video and everything is there's definitely more than one way to do anything. So any, there's no bad advice really that much. There is bad advice. Don't give that, but don't be afraid to give advice with saying that people don't own photography matter, anything. Like, everybody is great. You guys are fantastic. Thank you so much for watching my channel and my tutorials. It really means a lot. And I'm also going to be working on more tutorials and classes even soon. So keep an eye out for that. Be sure to subscribe and like check me out on Twitch as well. I mean effect underscore, underscore Tron on Twitch. I'm gonna do a lot more livestreaming tutorials and things like that. So I probably won't live stream motion tracking because it's a very timely process. But there you go. Popular joy. Thanks. Leave a like if you did and subscribe, I would really appreciate it. Thank you very much.