Monday, July 31, 2006
Last Tuesday, when I posted this photo:
showing 5 different versions of myself in one shot, Dani commented:
"i've always wanted to do this with my digi camera but don't know how.
can you help please?"
I figured if Dani wants to know, maybe some other people do too, so here's a little rundown of what I did to create the "Creative Clones" shot.
There are two ways to go with this. For the first, you have no desire to duplicate the result, so you just say "it's magic" and call it a wrap. If that's what you want to do, I'm totally ok with that (it's pretty much how I got through every math class I ever took in school) and you're welcome to skip all this stuff below and just scroll down to the end result or come back tomorrow or something! {grin}
The second route is for those of you who want a more mundane explanation so you can try to do something similar at home without a magic wand. Read on to see how that one works.
First, and most importantly, you have to have three things:
*A tripod to hold your camera perfectly still in one position
*A timed shutter release (or a willing helper to push the button)
*Photo-editing software that allows you to work with layers (I used Photoshop Elements)
If you have all those things, then figure out where you want to stage your shot and set up your tripod and camera so you have a clear view of everything you want to include. For the Creative Clones shot (hereafter referred to as the CCS, to save some typing!) I set up the tripod just outside the door of my workroom, making sure the view on the camera screen included my computer desk and at least the beginning edge of my sewing table.
If you want to use props, including changes of wardrobe, it's good to organize those ahead of time too and have them outside the area you're photographing, but close by. It saves time in the long run.
When you have your clothes and props ready for one version of yourself, set the camera's timer and push the button and then go get into position. You might want to do this more than once in case you move or blink or something at the wrong moment. Once you've taken 2 - 3 photos in that position, switch position (and clothes and props if you want) and do the whole process again. Lather, rinse, repeat until you have 2 - 3 photos of yourself in every position you had in mind for the final shot. Then load them all onto your computer and look through and choose your favorite for each area and you're ready to start the photo editing magic!
I didn't save any in-progress photos of CCS, so I staged a much-simpler one yesterday with only two positions and no wardrobe change to show you what I mean. These are the two photos I decided to work with (small versions of them):
(not exactly flattering, but Oh WELL!)
Determine which figure is going to end up being the farthest back in the final photo and that's the one you use to create your background - your "canvas" to which you add all the other figures. You don't want to work directly on your original though, because sometimes you need to go back and fix things after you've been monkeying around for a while, so the first thing you do in the editing software is to create a duplicate of whatever photo is going to work as your background and the copy becomes your canvas:
(I know these photos are crummy...sorry! I should've used screen captures. Next time!)
Next you can open the file for the second figure. I don't bother duplicating this one because I can always undo any changes to it without affecting my work-in-progress. In this photo, if you look at the thumbnails at the very bottom, you'll see I have two versions of the first photo (the original and a duplicate) and one version of the second photo:
In looking at them side-by-side I can see where I need to draw lines to select part of the second photo to move over onto the first photo. When possible, I try not to draw around just the figure because you end up with really harsh, unnatural edges, especially in areas like the hair. So as you can see from the photo below, I mostly drew lines that selected the room and all, only closing in and drawing around the figure in areas where one figure will overlap the other, such as the arm and hand of the second figure:
Once I've selected that area, I just drag it onto the duplicate version of my first photo and line up any overlapping edges as cleanly as possible. In this photo I only have to do this once, but for the CCS I did that step four different times - once for each of the extra figures going into the final shot.
Then it's just a matter of doing the fiddly bits like cleaning up any lines that don't quite match perfectly. This is why the tripod is so important. The more still you've been able to keep the camera, the less of this part you have to do!
This step is also why it's important to keep a "clean" version of each photo, so in case you've gotten overzealous in how much of the room you've moved along with a particular figure and accidentally lopped off someone's head or leg or something (photographically speaking, of course!), you can go back to the original version of the mangled figure and draw a whole new shape around just the area that got messed up and slide that over onto your group shot to hide the BooBoo.
Once you're happy with how everything is arranged and satisfied that everyone has all the pieces and parts they're supposed to have, you can flatten the image (combine all the various layers into one image) and do the final touches to the image as a whole.
This is where I do things like use a smudge tool to ever-so-slightly blur any edges that look too harsh and "cut out" to me, or to blend edges that don't quite meet seamlessly. It's where I use the paint tool and a fairly transparent setting to add any shadows that should be there, but aren't. You can see that in the shot below if you look below the arm of the me that's standing - I added a shadow to the me that's sitting because there would have been one from the flash if there had really been two of me.
And that's about it - the bare bones of it anyway. The rest is just practice, Practice, PRACTICE!
Here's the result of this more simple version, which is also today's Daily Art Thang:
"Telling Myself What To Do"
<< Home