Mark Behm has a fantastic step by step of his process over at Mark’s Sketchblog.
I love his approach of starting with a silhouette, doing everything in black and white, painting in the light, constantly flipping it to maintain proportions, using clipping masks and then flattening, and making a color comp and then sampling from that. It goes a bit fast, so I had to watch it a few times, but I like how it wasn’t 13 minutes as originally intended.
So here’s the last one. By this point I was feeling much better about my manipulating abilities, and I really want to start working on more mattes again. I’ve taken a bit of a hiatus from matte painting, honing my compositing skills and whatnot. I have done some much more professional yet also more subdued mattes for City of Ember. They’re almost completely hidden, so it was nice being in the lime light for this much smaller production. I can say there’s at least a minute and a half of this movie with my mattes as the background, and I really appreciate the way my coworkers and boss were able to bring out the best in me for these even if they were all done in a fairly restrained time frame.
Image Given. I was told to turn it into more of a jungle from 100 million years ago.
This attempt didn’t include enough of a hill or give room for the T-Rex to stomp through so…
I ended with this. I really had my process down by this last matte, although looking at it now months and months later the colors and tone seem far too uniform. What I’m getting at is I need to start doing more of these cause I think I might have a knack in this area.
Now I’m the first to admit I have a long way to go on becoming a photorealistic painter, so when my boss came to me and decided I was ready to handle creating a dino skeleton I was both honored and terrified. It seemed the challenge escalated with each matte, so by this later one I was feeling quite confident. I said, “of course I can do that” and gave it my best. Next I need to get off of this photo crutch. I still have so much to learn
Image Given.
Here’s what I came out with. For all of these I ended up using Creative Common images I found on flickr. Next time I do this, I’m going on a field trip and shooting my own ref. It will be more my own, more fun, and I’ll have far more control. I don’t even think this way is cheaper, for the time I needed to spend making the found images work. With that said, I think this turned out well.
These were fun because it started to get into what most background artists work with-namely combining some photographic, some painting, and some 3d. I was given a 3d model, and did my best to combine it. Nothing like spending time at work searching on flicker for pictures of bloody wounds that I could merge into a dinosaur.
The before image. Clearly there was much work to be done.
And the finish. In the final movie they ended up doing a zoom out from this image, which I think worked nicely but didn’t give it as much screen time. With the 10 second holds they had on some of the other mattes, I could live with that.
I’m going be honest. Nothing screams awesome more then getting paid to paint in bloody dinosaurs. It was nice comparing notes at the end of the day with my girlfriend who raises funds for homeless shelters.
“So what did you do today?” “Oh, me? I worked to get funding for a new program that would help over a 1000 families a year get back on track. It’s looking like it will go through. What about you?” “I painted bloody dinosaurs.” Yep. I win. Dinosaurs trump all.
The before image. I was told-make it dead and little else.
So I came out with this. Then I was told it was killed by an arrow, not an animal attack so I had to scale it back a bit.
I had Andrew reposition the model, and ended up with this. Clearly not my best work, but good fun all the same. It was held on screen for far longer then I thought justifiable, but who am I to argue?
I was really happy with this one for two reasons-first, I thought it ended up looking rather nice. More importantly, if memory serves, I did this towards the end and was able to pull it off in less then a day and that’s where the improvement really showed itself.
The before image.
And the after image. I really wish I could go more into my process of how I did this, but truth be told-it’s been nearly half a year since I worked on it and I really can’t remember any of the details. Damn nondisclosure agreements until the movie comes out…oh well-just enjoy the before and afters, and use your imagination for process.
There’s seven pictures, so I’ll split this into seven posts.
The before image. If memory serves this was the first matte I worked on. I was still figuring out what I was doing with this one.
I got this done, and then went on to the next matte. I knew this was no good, but for the life of me, couldn’t figure out what else to do with it. As I said, this was my first stab at it all and I still had plenty to learn (for that matter,I still do)
So I came back to this image having to put in a path, decrapify the river and make it work a bit more. I came out with this, still not completely satisfied.
When in doubt, smother with clouds. Between that, and fixing the colors, this got much, much better by the time I pronounced it finished.
I thought I’d talk a little about the mattes I did for the movie 100 million BC. I spent a lot of time working on these, and they’re the first successful photorealistic pictures I’ve done. This first one I was supposed to make it look like a giant drop down.
The before image.
This picture took far longer then what I ended up going with. It was decided it should be more of a drop and less of an angle so I ended up completely scrapping this.
This final image ended up using reference my boss shot in Hawaii with a couple of clouds slapped on top and some greenery to bring it all together. The final image took less then a day, but the inbetween variations took much longer.
