Smart Objects and Composites
What you're gonna notice that happened to this, and the reason why I'm showing you this is because you know what happened to this one over here, how the edges got blurry when I did that Gaussian blur? I wanna show you how to fix that before. So we'll go ahead and close this down, open this up. Okay. So what we have here is basically just a regular gallery-style look. We can fill these in with whatever pictures we want, so let's say you're trying to showcase your work on your website in a clever way that's not necessarily just some slideshows, and you wanna make it look like your work is in a gallery. We can go ahead and fill these blocks in with our work, and you're gonna see that these ones, that's kinda easy, okay we could do that pretty easily. That's kinda easy, that's kinda easy, but look at these ones, they have perspective attached to them. So the reason why that last smart object layer kinda failed us and got blurry on the edges was because we made that shape a smart object at ...
the wrong time. There's a workflow to this and when you want to make that shape and turn that shape into a smart object. So if I were to go ahead and make a shape on here, so let's go ahead and zoom in here. I'll make a shape, just like this. If I wanted to make this fit exactly into this space, and I started pulling these handles now, at this point, and then turned it into a smart object, it's not gonna remember that at one point this was a perfect square. It's just gonna remember what you told it to remember when you told it to remember it. So if I go ahead and right-click on this, and Convert to a Smart Object, and then I zoom in here, notice how it wasn't a perfect selection for that area. If I press Command or Control + T, and I press Control on these handles and move this to here, and move this down to here, and then move this here, and then here, Didn't seem like there was a whole lot of perspective going on with that area, did it, but there was. And I'll press Enter and commit to that. What happens is because we've converted that to a smart object before we did the transforming, so Photoshop remembers that. Photoshop remembers that at one point, that was a square. Now, with this one it might not be easy to wrap our head around, but let's do this. Let's do this on this one right here. Let's make a shape, any size shape will actually do at this point, that'll be fine, right-click, Convert to Smart Object, Command or Control + T, and then, as I press Control, I can fix that box there, Control and fix that box there, Control and fix that box there, Control + click down here, move this up, and fix that box there. Now, this frame does go on, doesn't it? Goes on a little bit further, so we can even do that to, if we zoom out here, if we press Control and space bar, and then Fit on Screen, and then Control and space bar and pull to the left, I can even drag this out a little bit, to there, and then press Enter. Because what Photoshop is gonna remember, Photoshop is gonna remember that that was once a rectangle that was this big, but then you converted it into something that is now a little bit wider. So if I were to go ahead and double-click on this, that's the rectangle that we started with, to make it fit into that space. So if I click on a photograph to put in there, let's just put this photo in there. You'll see how this is gonna warp it, we press Yes. Notice how it cuts off the sides and it fits it right into that perspective of the image, because it remembers that that at one point was a rectangle that has now been warped to fit that shape. Same thing here, if I make a shape that is relatively the same size as this, press Command or Control + T, I'll zoom in a little bit more, Control + T, to fit those edges. Okay. You know what I didn't do first? I didn't turn it into a smart object first. Command or Control + Z, let's go back, right-click, Convert to Smart Object, Control + T, and then we'll do this, and there's another thing I wanna incorporate in this one that's gonna make this a little bit more fun, and that's gonna be layer styles. So I'll make that about that size, perfect, press OK. Double-click inside here, so I can change this to whatever I want it to be changed to. Let's just use this photograph, move it into that rectangle, exit out, save it, Yes. Right now it's fitting, it's fitting into a perspective, it's changing the perspective of this image as it puts it into that rectangle, but it just doesn't look like it fits, does it? It looks like someone just put an image inside that canvas, and there's no gradience or anything that looks like there might be like a glare on it. So if I double-click on this rectangle, and go into the Layer Styles, I can go ahead and add a Gradient Overlay to that. Maybe I'll change that Blend Mode to something like Soft Light, and then change it to a Radial Gradient, and then a gradient that goes maybe from black to white. Press OK. I can move this gradient around, and then maybe increase the scale a little bit. Now it looks like it has kind of like a glare on it. If I change that radial from something like Radial to Reflected, again same kind of thing, looks like we have a glare coming from it, or it's a little bit lighter on the top than it is on the bottom. Press OK, a cool thing about this is because those layer styles were recorded into this as a smart object, if I double-click on that smart object to replace that image with any other photograph, guess what's gonna happen, it's gonna have those layer styles attached to them. So I'll take this image, let's just grab this one, and drag this and drop this into that layout, Shift + Alt. Save it out. Those layer styles are saved in there, see that? Pretty darn cool. So let me go ahead and minimize this. I'm gonna show you how this plays in resizing things for composites. So I'm gonna go ahead and open up an image. I'm just gonna open up this image. Now how this works when it comes to resizing images, especially for things like composite work, because I told you in the beginning, when you make things larger and you make things smaller, you're gonna be losing pixels and gaining pixels as you make that image smaller. So I'm gonna turn the background layer off on this, and I'm gonna duplicate it, I'm gonna go to History and I'm gonna duplicate. And on this one, I'm gonna right-click, and I'm gonna say Convert to Smart Object, and on this one I'm going to just leave it as a non-smart object. So at this point, I'm gonna press Command or Control + T, and I'm gonna make this smaller, much smaller, like this small, press Enter. I'm gonna come over to this image, press Command or Control + T, and make this smaller, much smaller, and press Enter. So if I go back to this one, because this was a rasterized image that I made smaller, if I wanna ever make that larger, watch what happens, if I press Command or Control + T, and I increase the size of this, look at that. We got our own Gaussian blur. If I go to the vector-based image, or what we made our smart object that then turned those rasterized pixels into what would be a vector-based similarity, press Command or Control + T, Shift + Alt, make that larger, look what happens. It remembered all of the data that was in that image before we made it smaller. So it is a good habit to get into using smart objects when you're resizing images, making images smaller, or more specifically, when you're doing composite work, like compositing a model onto a different background, turn that model into a smart object before you go ahead and put it onto the background, or before you put the background behind it, I should say, so that as you resize that model to get her to fit just right on that background, you're not losing that pixel quality as you make it larger and smaller, larger and smaller, larger and smaller. So we talked about quite a bit when it came to smart objects. We talked about what a smart object is, how a smart object works, and what we can do as photographers, and how we can use them in our workflow. We talked about smart objects from Adobe Camera Raw, how all we have to do is press Shift, and we can open that image as a smart object, and we can always go back in and redo any of those Camera Raw settings. We then used smart objects with filters, so that we could apply a Gaussian blur, and then go back into those settings, and even use the mask on that Gaussian blur. We then talked about smart objects as they pertain to straightening images, and we saw how effective that was on the vector versus raster. And we talked about smart objects with image layouts, and how you can change shapes, and apply images within those shapes. And then we talked about how to maybe make your own composite gallery-style wall, and how if you make it a smart object before you do any perspective corrections, those perspective corrections will be saved into that smart object. And then we talked about why it's important to use smart objects, specifically when it comes to resizing our images, because we saw that we just made a Gaussian blur by making a really bad rasterize decision.