Showing posts with label art from photos. Show all posts
Showing posts with label art from photos. Show all posts

Saturday, March 22, 2008

Maggie on Photographic References, Part IV

Okay, onto the last part of my changing photographic references for art series. I'm gonna answer all questions in a separate post tomorrow. Folks just joining can see the earlier parts here:

Part I
Part II
Part III

Tonight we're coming to my last two in depth mini-topics, both of which are actually massive enough to justify their own weekly series: combining two subjects into one piece of art and altering a photo to suit your own personal artistic style. I'm going to touch on these two items very lightly in the relative scheme of things, so if you have a specific question, be sure to ask it in the comments so I can get to it tomorrow.

So tonight I'm actually cheating and using the same example to illustrate both points.


COMBINING PHOTOGRAPHIC REFERENCES

I'm asked to do this a lot in my commission portrait work, and I'm sure it's pretty obvious why. Dog A and Dog B don't get along, but client would like photo of them snuggling like Brad Pitt and Angelina Jolie. Or Dog A and Dog B don't hold still long on their own, much less in each other's presence, and a nice photo of them together is about as likely as Brad Pitt and Jennifer Aniston getting back together. Anyway, whatever the reason, suffice to say that if you do portraits, you're going to eventually be asked to do a multi-subject piece. And you want to be able to say "yes." (Unless it's a portrait of Brad Pitt and Jennifer Aniston).

So let's pretend that these two pretty bad photos of Moose and Peanut need to be combined in some artistic fashion. We can't really just smush them together in Photoshop and tell them to play nice -- it won't work. At least not without considerable thought. And why, you ask, is that? A couple reasons:

  • Lighting. Notice that the light is coming from two entirely different directions. Not only will we have to adjust for that, but if we're putting them close together, we'll have to adjust for one of them casting shadows onto the other one.
  • Size. We're going to need to know what size they are relative to each other and adjust for that as well. A cat and a Jack Russell might not be hard to guess, but what if it was two mutts of indeterminate size? Or a giant-headed child and a midget? You need this information before you combine two subjects or you can end up with a real disaster.
  • Markings. You can just move around bodies wherever you want them without having photos of both sides of their bodies. Ditto with a person's face. You can't just turn them the other direction by flipping the photo. People's faces aren't symmetrical and animals' markings aren't either.
  • Altering body positions. Photoshop can only do you so much. If you're doing major changes, you need something a bit more powerful, which would be . . .

Your sketchbook! You knew I was going to say that, weren't you? I start out with two very quick value sketches of each subject as they stand. Just to get my bearings. Those suckers are on the left here and as you can see, they are very quick, very sketchy, and only have three values: dark, midtone, and light.



To me, these little ugly creations are worth gold. Do you see what I've established with them in only 60 seconds? The major shapes, the light direction, and what I would do to correct the value patterns (note how I simplified the background on Peanut, the terrier). Anyway, now I have my base line. I know what I'm really working with.

So I move onto my mock-ups. I can do as many of these as I need to until I find the composition I like. Referring to my value sketches and the original photos, I put the two subjects together. First I try Moose on the left and Peanut on the right. Notice how I put them nice and close. Artsy. Snuggly. Cute. I reverse the shadowing on Moose's face and give him a shadow from Peanut's face. When I execute this piece (that means "finish it" not "line it up against a wall and shoot it"), I'll have to know what the markings are on Moose's other side, if I use this mock-up.

Okay, I like that, but I want to see what happens if I swap their positions. Out comes the sketchbook and another value sketch appears. Again, quick. Painless. Just the three values and major shapes. Here's Moose on the right. I still need to adjust his lighting, but I don't have to play with his body shape and I can probably get away without too much shadowing on Peanut's face if I pretend my lighting is coming from a high right angle. I like this one. Looking at it now, I think I'd definitely make sure to knock back Peanut's big body shape there on the left back to midtone, because it's distracting as it is.

So you see how simple that was? Not at all painful.

Problem: Need to combine two photos.
Solution: Three Value Mockups, correcting for lighting and size discrepancies.

ADAPTING A PHOTO TO YOUR STYLE

This goes back to the conversation we had earlier this month about not slavishly copying photos but rather putting your own spin on them. For this reason, I get particular enjoyment from working from not so great photos, because that way I can take full credit for them turning out okay!

I thought a visual example would be easier than a written one for this. The two photographs you see of Peanut in this post were the basis for the two pieces of Peanut artwork also in this post. I think you can see how I took my style and pushed the envelope. Key points of my style are:

  • A lot deep darks.
  • Scattered hints of bright color
  • Intense saturation
The photos lacked all of those -- but I put them in anyway. Establish what you want the image to say and then let the photo work for you, not the other way around.

Problem: Copying a photo.
Solution: Use your sketchbook and your imagination to use the photo as just one piece of your artistic arsenal.

Next week's series, by the way, is on something important to me -- motivation, time-management, and artistic mentality.

Wednesday, March 19, 2008

Maggie on Photographic References, Part III

portrait commission/ work in progress
16 x 20" colored pencil on pastelbord
copyright 2008 Maggie Stiefvater.


I am interrupting a fun filled evening of child-rearing and art-making to continue my series on altering photographic references for art. (Last part can be found here)

Anyway, so I just rolled over an Oreo with my desk chair (remind me to tell you guys tomorrow about buying a new desk chair) and I'm amazed at how large a splatter pattern one Oreo can make when faced with light-colored carpet. Really, the government should study this. I'm sure it has some use in the defense area.

All right, last post I covered lighting and composition in photo references. Tonight I'm going to write about how I approach photographs that are lacking good color or nice backgrounds, two things that are particularly important in simple portraits.

CORRECTING FOR POOR COLOR IN PHOTOGRAPHS

This one is sort of easy. It takes a bit of leg work, but it's not really hard. I get photos with poor color all the time -- flashed photos often are washed out, dark photos can be brightened for details but often lack good color, weird lighting situations can create technicolor creations of terrifying hue.

In this case, it's a matter of finding another photo, or photos, that fill in the information gap. Sometimes, this will be very easy. For instance, I love the pose on the tiger shot above, one I took at the National Zoo last year (appreciate that the tiger was 8,000 miles away when I took this shot if you will). I know what color this tiger was because I was there to take the photo. However, since I didn't have a sketchbook with me to record the colors, another ref would help fill in the gaps in my memory. It's a simple matter to hit the internet (google's image search is a nice function for this) to find another tiger photo with better color. I'll use my pose and the color from the other photo.

In this case, it's not really necessary to track down a copyright-free photo, since you're not using the likeness or copying the photo in any way, but if you do need to borrow majorly from a photo in anyway, there are several places to find photos that are copyright free, like the WetCanvas reference library, or with photos available under the Creative Commons License, like Flickr (notice that not all photos on Flickr are available under the Creative Commons License, you have to check that box on the advanced search page).

Sometimes, however, the task is harder because it's a commission and you've not met the subject. Say it's a flashed photo of a girl with what looks like dark brown hair. You're pretty sure that it's lighter than the photo, but you're not sure how light. This is not the time to pull out hair dye swatches and start guessing. This is where normally I would send the client an e-mail asking them to search through one of the photo sources for a girl with the same color hair. I've not yet had a client be bothered by such "homework." I find that usually, however, I can just ask the client if they have a photo that has good color and bad everything else. Usually there's a blurry photo with accurate colors that they can send.

So.

Problem: Poor color in reference photos.
Solution: Fill in the gaps with other photos of similar subjects, with color sketches (if you made them), or with photos lacking in other elements but sporting true color.


ADJUSTING BACKGROUND IN PHOTOGRAPHS & COMBINING REFERENCES (part 1)

This is a biggie. Now, you'll have to pardon my example photos here. In the great Computer Porno Fire of 2007, I lost many of my photo references, and then in the Great Hard Drive Malfunction of 2008 I lost many more (for those who may not remember either, the first was when Norton AntiVirus downloaded 3,000 porno videos onto my computer into hidden folders - thanks, Norton! and the second was when Life decided it Hated Me), so I'm kind of running short of examples here. I'll have better ones on Friday, promise.

Okay. So what do you do when you need to change backgrounds? Let's say you're doing a portrait of your charming daughter and the background is either boring or shows something really unartistic behind her, like your neighbor's dog making love to a sofa pillow or the such. The simplest way to change this is to do a studio background, just a single color, either flat or in gradient. But I hate studio portraits with a passion (my apologies to those who love them) so for me, I need all of the image to be doing something.

So let's imagine that I have consumed an ungodly amount of sweet tea and I want to do a portrait of my daughter as if she is in the middle of this lovely field near Warsaw, Virginia. I have two real options for references.

When I was first starting out, I'd use option A -- I'd digitally combine the two photos in Adobe Photoshop. This is problematic for a few reasons. While it gave me a reference instantly and beautifully combined, it didn't require me to use my artistic brain. So if I'd just thoughtlessly combined the two photos above, I would've been making a major lighting error (note the direction of lighting in each) and I would've been forced to make all my position and sizing changes dependent on how well I knew how to use the program. Any color changes I thought I'd needed I'd have to tweak individually. There are entire college degrees based around this stuff, folks.

So, as I got better and starting using my sketchbook for quick 2 minute preliminary sketches, I started using option B: sketching the various options in quick 2 minute sketches. This is the option I still use today. It's much faster for me than Photoshop, and it also sorts out a ton of problems a lot more effectively. I'm not so worried about details or likeness in these little quick sketches -- more about how values and shapes affect each other.

So enter my ugly two minute sketch. Is it great art? No. Did it work out a lot of problems in my head for this piece? Yes. Notice a couple things about it.




  • Light source is identified.
  • Possible color conflicts (similarity between sky and shirt color) are identified and a possible solution (intensifying saturation on shirt) is tried out
  • Two different crops are tried out. If I was really doing a piece like this, I'd probably have three or four little sketches like this, the others in black and white, trying out landscape versus portrait formats, having her standing in the grass versus up close, more of her face versus more of her body, etc.
  • Another advantage? A lot of times, portrait clients are very happy to get these sketches along with their portraits, if you're willing to give them up. People are fascinated by the process.
Now there is a compromise between a flat studio background and a fully detailed one -- it is possibly to just simplify and blur the background of photo you already have. In this case, there's a great exercise you can do to figure how to blur elements. Focus on something -- like my interesting blog post on your computer screen -- but make yourself away of the way you're perceiving things outside of your direct focus area. See how uncontrasty and blurred they are? That's what you're going for.

Problem: Crummy background.
Solution: Eliminate the background, simplify the background, or sub another photograph.

Yikes. I just realized this blog post took me a stunning two hours to assemble, which means my portrait is going to be put on hold until tomorrow.

I should take the opportunity to plug myself, shouldn't I? If you're liking this series of blog posts I've been doing lately and feel like showing your support, please head onto cafepress and buy a t-shirt or mug with my art on it, sign up for a workshop, or e-mail my editor at Flux and tell him how brilliant I am. Okay, I'm kidding about the last one.

Next post will finish up photo alterations -- remember to post questions in the comments section. Next week's series topic will be one which is very important to me!

Tuesday, March 11, 2008

Maggie on Photographic References, Part I

"Lawrence" - 11 x 14" colored pencil on board.
Copyright Maggie Stiefvater.
Contact Chasen Galleries for purchase info.


(My version on left and reference photo -- actual size -- on right)
There has always been a major debate raging about using photographs as reference for artwork and this post is not about that (it's also not about player pianos, flocks of sharpie markers, or swear words -- which are also post-worthy subjects that may get touched upon this week). The only thing I will say about the debate here is that I do think that there is a place for photographs as references, as long as you know their limitations and learn to take/ use your own, and as long as you’re also able to draw from life. Drawing from life has an immediacy and freshness that I think you’d be hard pressed to get from a photo. And drawing from someone else's photo will always be revising someone else's vision rather than starting from scratch by yourself.

Okay. That aside. Onto the actual craft of using photographs for references, or more specifically, changing them to make a better piece of art. The more photo-realistic your finished piece of art is meant to be, the more thoughtful you’re going to have to be about these changes.

There are a few different reasons to change a photo, always because it’s lacking. It could be lacking:

- good lighting
- good composition
- good background
- necessary elements from another reference (like if you’re putting additional subjects in a portrait or putting a subject in a different surrounding)
- good color
- good detail
- your artistic style

Unless the photograph is yours and is absolute perfection, you will be changing at least one of these things. This is the point where I usually get two big questions:

1) How do I analyze a photo to even know if it’s good or bad?

and

2) How do I change it so that the changes look natural and realistic?

And here is the disappointing bit . . . because all I'm going to say is that I'm tackling these questions this week (with photographic examples! don't be too sad!) on Wednesday and Friday. If you have any other questions that this post brought up that I should be tackling as well, leave 'em in the comments. I promise I do eventually get to them . . .