Whip Up Your Own Size Collage

We’re all in to Size Fantasy, right? We’re all attracted to a notion that cannot exist under our current understanding of physics. To get around this we write stories, we draw pictures, and we learn programs to enable us to construct photos and videos that replicate aspects of the Size Fantasy experience. Each of these disciplines requires lots of practice, and some also call for expensive software and computers to cope with it. But the more you write or draw and the more you read and study others’ pictures, the better you get. You can always upgrade to a better camera, and you can buy a better processor. And you can wait for the editing programs to improve and get cheaper, even become free.

The process I’m going to describe isn’t perfect, it’s quick and hasty and it produces a good-enough image. Knowing what little I do about Photoshop, it skips the step where you can adjust the layers to more harmoniously match in terms of light and color, but there are ways around that. I’m entirely self-taught in graphic editing, using two powerful, free programs that let me shrink myself down and enjoy a much larger world.

RemoveBG is a clever program that recognizes the human in an image and strips out everything else. It’s still in development, but nine times out of ten it does exactly what you need it to. There’s a paid version that enables you to download larger images, which is good if you care about composing a high resolution finished product. If you’re just screwing around, you might as well stick with the free version (until you feel guilty enough to pay them what they’re worth). The advantage this program offers over masking in Photoshop is time and labor: it spares you all the detail work of erasing everything around your hair and fingers and folds in clothing, and it does it in seconds.

The other program I use is Pixlr, which is a robust (and free!) photo editing program. It comes in two versions: X offers an easier user experience, and E is slightly more powerful. You might as well start with Pixlr X if you’re a beginner like me. And if you’re a Mac user, I apologize in advance: I’m doing this on a PC laptop, and I don’t know how to translate the instructions for Mac users. Hopefully the differences aren’t too egregious.

Further, these instructions are only for creating the illusion of tiny people. Many of the techniques described are inapplicable to composing an image of gigantic people, and I have no experience with that. My apologies to my giant friends; you know I adore you, but this is my lane.

I’m going to walk you through the steps of isolating an image of yourself to impose upon a background, and how to size it up and tweak it to compose a good-enough Size Fantasy photo collage.

1. Stage the image

The easiest, lowest-tech way to fix lighting and make the shadows consistent between the two images (yourself and the background) is to take two photos in the same location, from the same direction. Make sure the sun/lights are shining on you the same as they’re shining on the setting.

Invest in a camera grip with hotshoe and tripod, or ask a friend to help you out, if you’re doing this outdoors; indoors, I have good luck with setting the timer on my phone camera and propping it in a wide coffee mug.

Perhaps the trickiest aspect to this is preparing the photo of yourself at an appropriate angle. If the camera’s too close, your features will look comically exaggerated—everything close to the camera will be huge, everything far will be diminished (“the Jack Kirby effect”)—and you won’t resemble your surroundings in perspective. If you’re taking a straight-on shot of a setting, the camera can be level with your head or body, but if you’re shooting down at the background, you have to figure out how to elevate the camera to look down on you too. If you’re shooting your setting at an angle, envision where you’d stand in that setting and take pains to turn yourself in a corresponding direction.

But we’re fudging this, so does it really matter? Play around with it and figure it out.

2. Isolate yourself

If you’re working with a conscientious helper, you’ll have one shot of the setting and several shots of you trying to get into position. That’s great: study your setting and select the best image of yourself, bearing in mind your expression, the lighting, your position, &c. Once you have your favorite image, delete the others so you don’t get confused.

In fact, start a folder into which you’re doing to dump all your rough images, and in that create two more folders. One will hold all your finished products, and the other will contain your assets: a collection of your backgrounds and, more importantly, more and more images of you in various poses with the background removed. You never know when those will come in handy. If you want, you can make a third one for storing inspiring images you find elsewhere and would like to work into a Size Fantasy collage.

Bear in mind: If you’re using a professional model’s work, the images of a pornstar on Twitter or Instagram, get their permission first if you intend to publish it anywhere. Their images are publicly available, but that does not mean they consent to being digitally manipulated into other people’s fetishes. If you’re going to keep these images to yourself, I guess no one can stop you, but the ethical and decent thing to do is politely ask a model if you can use a specific image and explain why. Most models will not notice your message, so you’re out of luck. Some will turn you down because they don’t understand or they’re creeped out: do not ask them twice. Respect their decision, thank them for their time, and walk the fuck away. A few will agree, and when you post the finished image, be diligent in linking back to them and thanking them again. That is how you earn esteem in a community and prevent people from warning others about you.

Open up the image of yourself, and use whatever graphics program you like to crop it down so you’ve eliminated almost everything that isn’t you. Creating a small graphics file makes for a much better isolated image, for the next step. Otherwise, you get stuck with a huge invisible background and a tiny little speck of yourself and it doesn’t matter whether you paid for the hi-res version. You’ll see.

Go to RemoveBG and upload your image. You don’t have to create an account if you want to use it for free. If you do buy a month’s worth of credits (40 images), you get the option to download a higher-resolution image, which looks nicer for detail work, but if you used sufficient lighting you may not need it. Also, the hi-res version just means it’s larger on the screen: the program doesn’t do better detail work. You may find you have to trim some extra elements or even restore parts of yourself. RemoveBG offers crude tools to do that, which is fine because you can do the detailed clean-up in Pixlr.

See what the Restore function’s doing there? It’s really easy to use, except the edges are too crisp, if you’re erasing extra artifacts around yourself. That will stand out because the edges around the image will be just slightly blurry, which is great for setting you into your background, but a crisp edge means more work later. Start with a well-lit image and an uncomplicated background, and RemoveBG will serve you well.

RemoveBG lets you download a PNG file with an invisible background and just the slightest haze around your image. It’s very good! Dig it out of your Downloads folder and move it to the folder where you’re doing your work.

3. Plant yourself in the background

Go to Pixlr X, click on Open Image, and open your background image. It asks you what size of image you’re going to work with. If you have a good camera, your background image will be large, and if you paid for a hi-res image of yourself (and if your computer can process it easily), go for the largest setting. Whatever else is going on, this will make a nice finished product. But if you’re not that invested in it, if good enough is good enough, use the other preset sizes. No harm in that.

Your background image is splayed across the screen. If you haven’t signed up for an account, you’ll get ads that cut into your workspace. Signing up is free, and it keeps track of your past work, so you might as well get an account. On my Windows PC, the function to add a layer is in the lower right: click on that plus-sign and upload the PNG of you on the invisible background. It’ll ask what kind of layer to add, and I think you can figure out to select the Image layer.

There’s that isolated image of you on the background, but it’s probably at the wrong scale entirely. With your self-image layer selected, grab the corner of your image and scale it roughly how you want it. Don’t worry, it will preserve the proportions.

I guess a pro-tip here would be to bring a tape measure and take a second background shot with the measurement, so you know how small to make yourself. Can’t believe I thought of that just now. I’ve been estimating it in my pictures, and I’m sure my scale has been wildly off from image to image.

Oh well! We’re just fudging it so who cares?

Once you’ve got yourself sized appropriately, zoom in to the image (Navigate, upper right) and enlarge yourself in your surroundings. The two tools you’re going to use now on the left-hand toolbar are Drawing (for the Erase function) and Retouch (for the Dodge/Burn function). On the right-hand toolbar, where your self-image is still highlighted, you’ll notice a menu button of three lateral dots, called a “kebab” (or just right-click on the layer). Click that, and now you can make your layer transparent. This is handy for when you want to position objects in front of you, to perpetuate the illusion that you’re really there and interacting in your environment.

On the left, click Drawing (looks like a paint brush) and click on the Erase icon. Size and Softness concern you now: a large size means you can erase more, but the softness means you get a wider aura of haziness on the edges of your erase tool. If you reduce the size you can do detail work, and if you reduce the softness you can sharpen the edges. At this point, try setting Size to 20px (use the slider or highlight and retype the number) and the Softness down to 10%. Or play around with them and figure out what they do, that’s cool too: you’ll develop your own preferences with experimentation and practice.

Simply erase the parts of you that are supposed to be behind any object that’s supposed to be in front of you. Be mindful of the hazy edge of the erase tool and use that to your advantage, since the large objects will also be slightly hazy on the edges. Try erasing with gradually smaller cursors, as detailed as you care to get, and remember to zoom in for precise work. Click on your layer again and return Transparency up to 100 and see how you did.

Is it absolutely convincing? Probably not. Is it good enough? Shit, yes!

Now, select Retouch (looks like a make-up brush) and go to the divided circle icon for Dodge/Burn; select Darken, obviously, because now you’re going to work on shadows. Also, click on the background layer (right toolbar) because you’re not creating a shadow on yourself, you’re painting it onto the background under you.

Shadows can be difficult, and at this point I won’t explain how to create a long shadow on a street, for example, when you impose yourself on a background and there’s an obvious light source but RemoveBG didn’t retain your shadow. You could use their Restore tool to keep it and then trim it in Pixlr, if it makes sense on the surface you’re “standing” on. Otherwise, yeah, it’s beyond the purview of this explanation to get into that.

There are three ranges for darkening: dark, mid, and light. Look at the big leaf I’m standing on/behind for this example. I created three streaks for each range, painting four times in place to show their effects. At Softness 50% and Strength 3, Dark increases the saturation and darkens the tint (“hue” is color and “tint” is shading). Mid does the same but more subtly. Light only deals with black-and-white gradients and imposes dark gray over the image.

Kicking the Strength up to 8, and only painting over them three times, you can see a much more dramatic effect.

As you get comfortable with these tools, you’ll use combinations of these effects to compensate for varied backgrounds and lighting effects. All you really want to do in most cases is insinuate a little shadow under your shoes to imply you’re there. Without that, you look like you’re floating or you’re a sticker someone smacked on a photo.

To start with, zoom in to about 320% (or just get real close), and guide your view to your feet and whatever you’re “standing” on with the slidebars or the little red outlined box in the Navigate window. Choose the light range, set your brush size to 10px, softness 50%, and holy crap remember to bring the strength back down to 3, and… just practice painting under your shoe, where you’re standing. Don’t paint around the perimeter (or do, and see what that looks like). There’s a dot or a cross inside the circle of your painting cursor: keep it inside the area of your shoe and keep the edge of the circle on or just inside the edge of your sole. Give one pass under the toe of your shoe, and try three passes under the lateral arch of your foot, right before the heel. You can undo anything you do, but try to be subtle (or don’t, and see what that looks like). Zoom back out to examine your work, too.

How did that look? Is it convincing? Is it good enough for government work?

Just keep playing with it. When the light’s directly overhead, maybe all you need is the the light range shading. If your shadow trails over another object, maybe all the ranges will help you replicate a shadow with color.

4. Finalize the image

You’ve done all the hard work, and you’re almost done. Here’s a little trick I like to use as a last step.

On the right toolbar, click on the layer with you. On the left toolbar, select the Filter tool (looks like a circle with seven dots in it). You’re going to blur your image just slightly to impart the illusion that you’re really there in the background. Zoom in on the objects around you, and scroll over to yourself: see the difference? A highly detailed image of yourself standing amid a much larger background doesn’t look consistent.

Overwrite the Blur setting to 2 or 3, and zoom out to see which looks best.

That’s it, that’s the trick.

If there’s still some annoying pixelation around yourself, feel free to go back to Drawing > Erase and clean up your outline. Just remember to reduce your cursor size and the softness, so you don’t look like you’re slowly fading away in parts.

When you’re ready for your last steps, click on the kebab of (or right-click on) any layer and select Flatten Image. All layers have become one, but you can undo this for the time being if there’s anything else you want to go back and change. If you were working with a huge image for the resolution but don’t want a huge image to display online, select the Crop tool (third icon down) on the left toolbar and click on the Select Aspect button. If you don’t know what ratio you want, peruse the Preset dropdown menu: it even has options for Facebook and Twitter dimensions, whatever that means.

The important thing is to crop first, then resize. Maybe it doesn’t make a difference, but I feel that retains the integrity of the resolution. Someone will correct me on this, I’m sure.

Once you’ve got your ratio set, grab a corner or a side, drag it around, and resize the crop to exactly the image you want to feature. The crop pane very helpfully outlines the “rule of thirds” guides, because sometimes it’s pleasing to have the subject of the image off to the side rather than dead center. When you’ve got it how you like it, click Apply directly below the crop settings. Do not click Save just yet, off to the right.

Now you can click the first icon on the left toolbar, Properties, and select Resize Image. I like my images to be around 1400 or 1600 pixels wide for social media display. Leave Constrain Proportions on, obviously, and click Apply when you’re done, and now you can click Save. You’ll have the option to save it as a JPG and choose the file size or a PNG. I don’t know what PXD is. If you keep your file size under 1 MB, it’ll look good enough for sharing online; bringing it down around 300kb or smaller will improve download time but introduce all sorts of ugly artifacts to your image. And remember to find the finished image in Downloads. Pixlr does offer instructions on how to save elsewhere, but they’re not specific to Pixlr: they instruct you on how to redirect where your computer saves downloads by default.

That’s it! You’ve staged your images and collaged your own Size Fantasy image! How’s that feel? Post it on your shadow Twitter account, start up an Instagram account, upload it to DeviantArt. Obviously you’ll want some kind of pixelation effect if you don’t want to show your face and reveal that you’re into this stuff: be mindful of your clothes and any identifying landmarks in the background, for the same reason. All that’s left is for you to grab your camera, go outside, and look for more inspiration, reexamine your living environment with new eyes. When you’ve practiced this a few times, you’ll discover you can knock out a size collage in under ten minutes, from posing for the photo to saving the final image.

This is a very low-level, beginner explanation of how to create a Size Fantasy photo collage. I purposely skipped certain aspects, like artificially creating a shadow on the background image, because that’s too complex to get into and, frankly, I’m not sure how. I mocked it up once. But if my instructions are incomplete, please let me know and I’ll improve them. Thanks for reading! I look forward to seeing what you come up with!

3 thoughts on “Whip Up Your Own Size Collage

  1. Very clear post. You’re getting better at these every time I see you post your work. Shadows are not optional if realism matters to the creator (and it should), so thank you for pointing that out.

    I don’t see myself using these programs at the moment, since I’m not collaging at the moment… but I do think that in the future I’ll plaster myself onto building or planet images, or the sizey like, so then this process will come in handy.

    (Although I can now select pixels from an image in minutes.)

    Thank you for the straightforward, instructive post! 👍🏻

    Liked by 1 person

    1. I think the process for imposing a giant person onto a landscape would require very different tricks with setting up the shot. A drone would be very handy, to get the city at the right perspective; otherwise, I guess a huge finger coming in through a window shouldn’t be hard to do. I’ll have to play around with that.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.