Layer Upon Layer Upon Layer

I first saw Photoshop when it came free on a coverdisc on a magazine and my friend Dion installed it on his PC, back in the early 90s. He thought that I, as an artist, would find it useful to create some of the artwork required for our little projects. But after opening it up I was mightily confused. This was probably version 2.0, and it did not come with a manual (before the Internet, before PDFs, physical books were the only way manuals were provided). I was completely lost by the multitude of buttons, tools, menus, and icons. Where do I even begin? So for the first couple of times I scribbled colourful swirls and then just sat there. I couldn't figure out how to do straight lines, the mouse had no easy way to draw like a pen, and so clumsily I moved the pointer around a few times, achieving nothing.

It was about five years before I opened up Photoshop again. I had just been hired as a website designer for the first time, and I was completely new to the specifics of that job. I was hired solely on my artistic abilities and my familiarity with computers, but had no specific experience with HTML or Photoshop graphics. So I was allowed about two months just to familiarise myself with both those things.

Now I had a purpose for using Photoshop. I was given a Wacom drawing tablet, and with a goal in mind could go through all the steps to achieve an image. How do I create a box with round corners and highlighted edges? How do I create a drop shadow, a gradient, or coloured text? This was version 4.0 of Photoshop, so certain things did not exist yet - no multiple Undo, no auto dropshadow and glow (known as Layer Styles), no Batch Actions. But I had a manual, I had a few excellent quality guide books, and a mailing list of Professionals including some Adobe employees, all of whom were really helpful.

And the key to using Photoshop effectively is understanding layers. Unlike drawing or painting, where you're stuck with a single sheet of canvas or paper to just pile stuff on top of itself, with Photoshop you have individual layers which you can move, erase, recolour, distort, fade transparency; an infinite amount of adjustment, for each individual small part that you choose. It's a fantastic way to create.

Over the years I've learned a lot more about Photoshop, that there are multiple ways to achieve the same thing, that each subsequent version adds features that make some jobs a lot easier, it integrates better with Adobe's sister-products, and evolves and adapts to be useful for new technologies.

And one day I had to learn to use another of Adobe's products, called After Effects. When I first used it, I didn't understand just how powerful it could be. It seemed like a fewer-featured Photoshop. But in fact its role, to composite different video elements together for Visual Effects, requires such specific tools that comparing it with Photoshop is futile. After Effects is its own thing, and arguably even more powerful.

But one similarity, that it uses even more effectively than Photoshop, is its use of Layers.

I am slowly working my way through each shot for my Steampunk short film, "Eleanor Xandler: Temporal Detective," mostly in chronological order, and have just completed one of the most complicated shots of them all. And that's because it's a shot I didn't actually film on the shoot day. I realised, after looking at the rough edit, that there were a couple of shots that I really needed in order to transition from one shot to another effectively, and I had not thought ahead enough to have filmed those shots, so I wondered if I could make them up using what I could find amongst the clips I had.

One particular shot was from after I had called "Cut!" and the actress playing Eleanor, Sarah Breen, looked over to me to check if I was happy. That glance was all I needed to convey what was required, so I had now found my missing shot. Simple. But there was another shot I needed that was vastly more epic, and luckily this is where shooting on greenscreen, and having plenty of Behind-The-Scenes photography, each come in handy.

I wanted an Establishing Shot, one that shows the audience the location where the scene takes place, and orients them to where the characters are standing in relation to each other. I hadn't thought to film such a shot, because in my head I knew where everyone was. But that's not good enough, because of course I know, it's my story. I know plenty of things that the audience do not, it's my job to make sure all of those things are conveyed where necessary, be it through dialogue, performances, or editing. So now I had to create a shot that told everyone where my two characters were actually standing.

And where they're standing is in a huge dark smoky warehouse, filled with junk and decorated by 1880s style structural architecture. I had to create an over-the-shoulder shot of Eleanor entering through large doors, pointing her weapon at the Professor who was at the far end of the building, all while showing off how large and beautiful the warehouse was.

The warehouse 3D model had already been built, as an exterior and in preparation for all the interior shots that I was going to composite, but only in a limited fashion. Only build what you need, as anything more is just going to slow down the process. Well, now I had to expand on that, as this additional unplanned angle meant more of the warehouse would now be seen. I added in extra features, including a spiral staircase, an extra wall, a couple of piles of junk, and additional detail on the interior skylight roof.

During the shoot I had a B Camera, which is a second camera recording the shots from different angles and framing, giving me extra footage I can cut to should I need it. One of those shots was almost an over-the-shoulder of Eleanor, except from the wrong side, and that section of it was only 30 frames long. I needed 100 frames at least, but I clipped the shot out, flipped it to match the angle, and then stretched and reversed the shot a couple of times to extend its length. This can sometimes be glaringly obvious, I've seen reversed shots in feature films before and there's a few physics giveaways that you can spot sometimes (a turn or a blink in reverse looks slightly peculiar) so in order to hide the trick I made sure the part of the shot that was Eleanor would be completely out of focus with the depth of field.

In the far distance is the Professor. None of the shots I had of the Professor were full length, they all cut off at his knees or higher. Extremely luckily there was a single photo taken amongst the Behind-The-Scene images which included the Professor's legs, so I grabbed that, flipped and scaled them to match, and then animated them to match his movement, which wasn't much but still had to be done.

Unfortunately the angle of the Professor shot didn't match the perspective I needed, which was most obvious with the Console Panel he was standing at. So I had to replace the console with another BTS photo, which meant I had to cut around the Professor's arm (a technique called Rotoscoping) so that it looked like he was manipulating the controls accurately. The side of the console prop also has no surfacing, so I had to fill in that blank as well.

The warehouse is full of smoke, steam, and dust, because this is Steampunk and that's what the goggles are for, so I had to layer in not just multiple instances of atmosphere, just like if there had been a smoke machine on set, but I also had to include the shafts of sunlight that came through the door and skylight.

Couple all that stuff with the camera move, which is a pan down from skylight to Professor, that I had to track and match, it makes for a very complicated shot that should, in viewing it, go by quickly and seamlessly, and convey exactly what I wanted, that the space of the building is so big, and this where the characters are standing.

Piece By Piece

I made another babe-related wallpaper today, and I thought I'd show how I made it.

To look at it, I hope that you'd think it was a genuine untouched photo. That's certainly my plan most of the time. Whenever I make a matte painting, I try to get it as seamless as I can, so it's sort of become my default approach, though I don't always achieve it (I am notorious for cutting corners and cheating).

So I started with this original photo of actress Missy Peregrym (from the TV show Reaper, and a brief role on Heroes). If you look her up on Google Images, this is the pic you'll find most often.

But there's a minor problem with adapting it directly into a wallpaper - it's in a profile orientation; that is, it's taller than it is wide. I could crop it to fit, which I did try, but it cut off too much of her lower half, which was a great shame; plus, she's so centralised in the framing, it was not as nice a composition.

I took the highest resolution copy of it I could find, kept it full length, and scaled it to the vertical height of my monitor, which is 1050 pixels. Then I expanded the width to 1680px, leaving a huge block of empty space to the right.

There was a second pic I found online, from the same photo shoot, which had a little more of the door behind her, so I knew I could match that up if I distorted it to align the perspective. This filled up about a third of the empty space, but still left a lot more to fill. So I took that same pic, flipped it, and, after more distorting, joined it up like a mirror image. An overlap and a soft edged eraser blended the mismatched edges.

Because simply flipping didn't match exactly, I extended the edges of the panels with the clone stamp tool, maintaining their angle. Then I deleted the light on the wall, eliminating the obvious repetition, and mixed in a dark gradient to keep the corner in shadow. That not only drew the eye to the lighter part of the image, where Missy is standing, but also helped hide the fact that there's no door handle. I bet you didn't notice that.

I removed the word that was painted on the door, and added Missy's name there instead.

But a couple of days later I was looking at it again and I realised a couple of things. One is it had a horrible green tint to the image, presumably as a stylistic choice. I am red-green colour blind so it wasn't as obvious to me as it might have been to others. I have adjusted the tint back to a warmer red.

Then the other thing I did was changed the opposing wall. As it stood, it looked like a door at the end of a narrow corridor, but I think it was more likely to have been a corner of a larger room, so I took the rail and skewed it to be straight, and re-textured the wall to remove the corner-ish nature to it, so now it looked like the room was bigger.

And that's how I did it.

It was a relatively easy pic to put together, too. Some that I have done in the past have been ten times more complex than that. Always fun, though.

Related articles by Zemanta

• Peregrym: 'Heroes end would be a bummer' (digitalspy.co.uk)

Virtual Environments

The main role for a digital matte painting is to create an entirely virtual environment, which is to say, a completely artificially assembled background for a shot.

They used to be made using paint, often on glass, which is then blended optically, that is practically, and re-photographed. But these days you can use photographs, blend them together seamlessly, and then apply them directly to 3D CG models to give them a small amount of dimensionality.

The art of putting all the elements together for the final shot is that of the compositor's. Though I have to be familiar with compositing as part of my job, it's not quite my forte; I am better at dimensional-related things than I am at colour matching and rotoscoping (clipping around edges). So the compositing for Checkpoint is being handed over to its Director, Rob. My job is just to create the background plates.

One of the pickup shots we filmed a few months ago contains possibly the most creative camera move of the entire shoot - a dolly and pan combination. The camera moves rightwards sideways along a track, simultaneously the camera pans left, while the actress runs up to camera, all in front of a greenscreen, which we set up in Rob's backyard.

The usefulness of a greenscreen is simple in effect, but it's not until I started actually doing all this work that I have come to realise just how wide the potential really can be. We can film anybody in any clothing and doing any activity, and then place them in any environment you can think of - in a French village, a Cafe, on a roof, off the side of a skyscraper, in a sewer, on the Moon... or on a country road near a UK airfield.

What do country roads near airfields look like? Well, almost anything we want it to look like. I decided on a stony road surface, a length of grassed verge, and then, 30m beyond, a line of pine trees. I figured I could put that together, using photographic elements, combined to look photoreal. Then I apply that onto 3D geometry, each plane representing a textured surface. Simple, but it has a limitation.

In my software, the only way for that to be useful is if the virtual camera stays static. The projected image is always oriented from the camera's point of view. That means if I move the camera, the image will move with it. This is no use, as I want the image to stay locked onto the geometry, not slide along in relation to the camera.

There is a solution, which is to check the box that fixes the image, relative to a specified frame of the camera motion. For example, if I set the camera up on frame 1 to point where the image best matches the geometry, and then set it to be fixed there, then, when I move the camera, the image will stay locked at that orientation, even though the camera is in a different place by frame 257. But what that means is the camera then leaves the image behind, it disappears out of view, and now you're pointing at blank geometry.

But there are solutions to that, too. One is to orient it to a different camera, perhaps one with a different frame scale, to give you more space to move your proper camera through.

Another solution is to use the tiling feature, which lets you repeat the image in any axis you wish. But that means your matte painting image has to be seamlessly tileable, and with its geographical features oriented in a parallel fashion to keep the tiling logical.

And this is what I did. I included four horizontal stripes - road, grass, trees, and sky - and blended the edges so they repeated seamlessly. Then, when the camera moved and panned sideways, the landscape continued horizontally along the geometry, and, though it repeated, it wasn't too glaringly obvious.

The repetition can't happen on all axes, so the top of the image, the sky, needed to be blended. I repeated its edge colour infinitely, so the "sky" continued up as a solid shade of blue.

And though that's the crux of this shot's makeup, it doesn't take into account the hand tracking of the three dimensional move (because the tracking software didn't fnd enough points to figure it out), the car model that is sitting on the roadside, the particle-created grass blades in the foreground, and all the other doodads that make up variation in the landscape.

I love creating something out of nothing.

Checkpoint: Hadrian's Wall

Sometimes you never know just how complicated things can get.

Whenever I'm called on to do a visual effects shot for someone, my normal reaction is to say "Sure, I can do that" and about two thirds of the time it's true, I know exactly how to do it and that I'm capable of it with my current level of skills and software. But the other third of the time, I am secretly thinking "I have no idea how to do this!" and will rush out to do some research to figure some things out - can the software do it, do I need additional resources, am I really able to do this, how complicated will it be?

There's a shot in Checkpoint that I've been working on, off and on, in various pieces, over a period of six months. It's a simple shot - we filmed our hero on a bridge over a railway track, and had planned to do a crane shot over the wall of the bridge showing a train on the tracks, with a railway gun. So my original plan was to comp in the train, maybe replace the gumtrees with European pine.

Our attempts to do a crane shot handheld were a bit of a disaster, we really ought to have thought it through a bit better and borrowed a crane to achieve the shot. And we ought to have picked a day with nicer weather, too. But our hands were tied, as is so often the case, and we got what we could, as best we could.

So now it's my job to fix it.

The train model was already long completed, so that wasn't a problem, though it did take me a long time to build.

The first thing we knew we had to do was stabilise the shot as we follow our hero crawling along the wall, because it shakes all over the place. The problem is, with stabilisation comes distortion and motion blur. We can't really do anything about that, though, so I have reduced the shake down gradually as the shot progresses, so that the distortions are minimal.

Next thing we knew was that we would need a 3D representation of the railway line and valley, which had to be a matte painting including distant fields of the French coast, a complete replacement of the trees with European pine, and a French village. So I assembled the elements into a 2D matte image, using the best shot we had of the railway line, but it wasn't good enough for the final shot, and we knew we had to go back to the location to photograph high resolution images of the valley.

The day we arrived was a beautifully sunny day, and the shots of the valley we took were great, but as it turned out, not perfect, because as we crane over and tilt the virtual camera in the 3D environment, it distorted the verticals and all the trees looked like they were leaning over. The answer was to go back again and re-photograph the valley from a slight angle, so that the distortion would be eliminated, achieved by having the final angle of the shot match the actual angle the image was taken, and not fake it like the first try.

One of the issues I had with the crane move through the valley, though, was getting the sense of distance for the French coastal part of the shot. In reality it was a photo of a farm in Holland, but I had it comped in and needed it to move relative to the camera so it felt like it was miles away, and not a flat image. This meant finding ways to have the fiddly crinkly edges of the trees blend in with the background without distorting or obscuring it as it moved. So I had to mask them out using alpha masks, careful to keep the bleeding edge suitably coloured to blend, and move all elements to give a sense of depth. Rendering each layer separately was the only way to achieve this.

The good thing with this approach, though, is that I didn't need to use a tracking program to recreate the environment, and 'lock' the train model onto the real tracks. Instead I am able to apply the train actually to the virtual environment, because they're both 3D models in the same program, and are fixed entities. That is a relief, because 3D tracking is still not perfect, and is a bitch to get right.

The next thing I realised we needed was the wall of the bridge. It isn't enough to use a 2D image and animate it, I needed to have a 3D representation of the wall so that the depth as we craned over it felt realistic. I tried to find a way around this, because I was afraid it wasn't going to work, but in the end it was unavoidable. I built the wall, and the uneven rocky surface it possessed, and applied the textures of the real location that I took pics of. The end result is the wall looks completely convincing, and the crane move is seamless, so it was well worth it.

However, big problem. The original footage was taken on a grey overcast day, and the other photographs we took at later dates were in glorious sunlight. They don't match. I have to apply this footage onto the surface of my wall, while the virtual camera watches it and cranes over it. If the footage and the wall photos don't match exactly, it will look wrong.

I had to alter the colour of the wall image so it matched a grey day, then mask around it with a soft edged blend to overlap with the original footage, so you cannot see the mix between them.

Then I have to render the bridge part of the shot separately to the valley part of the shot, so that when Rob colour grades the two, he can either reduce the vibrancy of the sunny day valley, or increase the vibrancy of the grey day bridge, so that they blend seamlessly. I also have to provide him with an alpha channel mask of the footage so that he can separate the two shots when he grades them.

The train has to be rendered at high resolution. Soldiers have to be added, probably digital ones because we don't have any footage of live ones. Steam coming from the train, smoke from one of the stacks, atmospheric haze, and birds flying, all will add life to the finished shot and give it the energy to make it convincing.

It is the most complicated thing I have ever done. So many elements, so many techniques, so much preparation, and yet every single part of it I was absolutely confident about being capable of achieving it. Individually, anyway. If I'd realised just how complicated I was making it for myself, I would've come up with a different idea for the shot and convinced Rob to do that instead.

Like hire a real train.

These Ain't Your Grandma's Matte Paintings

Matte Painting is an astonishing art. The real geniuses of this art of the last 100 years of movie visual effects are true masters. Using nothing more than paint and their eye, they could match colours, create photo-real imagery, that matched the live action seamlessly, all done effortlessly. It's bewildering what they're capable of.

These days that art is no longer practised in the same way. Computer graphics and digital cameras means we can now use photographs to assemble together, to make our background plates and matte paintings, eliminating the practised art, and replacing it with reality itself.

In many ways it's disappointing to lose that specific approach to a relatively young art. But on the other hand, there is still a good amount of skill and artistic endeavour involved in digital mattes, especially when creating a fantastical environment that cannot exist in the real world.

I haven't had to make anything too spectacular before now, most of the time it's been a "real" location. Augmented to suit our needs, true, but basically made up of tangible and familiar elements.

Click the image for large pic
And today's work was just such an effort. I had to make an airfield runway from a particular close angle. It shouldn't have been anything too fancy and modern, because it was set in 1940s wartime, so it needed to look suitably worn and well used.

I found a whole slew of pics that someone had taken of a WWII German airfield, that has long since been abandoned and disused. It looked overgrown, but it had a suitable length and flatness to it that worked perfectly.

The land around it was a little crowded with new trees that had started to fill up the unused empty field, and the photos were all taken on the runway itself, instead of to the side where the virtual camera would have to be, so my job was to blend the photos together in such a way as to eliminate the trees and replace one side of the runway to make it look like we are standing in the grass next to the tarmac. Coupled with having to clear the overgrown weeds, it was a little bit of subtle work.

I may have to green up the finished image once we go full resolution and finalise it, but until then I believe I have figured out the right angles and created a suitable matte painting for the shot.