35milimetre

View Original

All You Need to Know on Photography Composites

When I first installed Photoshop on my PC, my initial intention was to do things that could be done in a dark room. My flat at the time didn’t have enough space to set up an actual darkroom. At the time, I thought that I could do most of the things I could do in a dark room with a film scanner and this software. It didn’t take very long for me that in a digital “darkroom,” a person can easily surpass what is achievable in a conventional darkroom.

Now fast forward 21 years; I do freelance photography editing work (I like to call independent visual post-production work); I receive all kinds of requests. Most of the time, photographers and ad agencies have a good grasp of what is achievable in a digital darkroom, but now and then, I receive requests next to impossible. I’m not counting the demands we see on TV series or movies, which I can summarize as enhancement requests. Of course, these are not possible; if an image is very blurred, we can’t just make it camera sharp. I would have told you that it’s impossible though Google has some papers, and they are working on AI solutions to remedy such issues.

One of the things that you can do with Adobe Photoshop is photo compositing. When I first realized what is possible, I was fascinated, and like all editors, my first composites were terrible. Photo compositing is a subject that I’m in love with both artistically and commercially. It’s one of the subjects that opened my eyes to a new world and still fascinates me. As an example, I love the works of Erik Johansson, who is a master in artistic photo compositing, and I hope I can hang one of his works on the walls of my home. I have been composting for a long time, and compositing offers tremendous advantages to commercial photography, helps to decrease productions costs of photoshoots which is one of the priorities on almost all photography projects. Another advantage is the ability to do the impossible or seemingly impossible, just like Erik Johansson’s works.

As humans, we are awake from 14 to 18 hours a day and continuously see things during this time. As a result, our brain is highly trained in detecting the “fake,” even though sometimes we can put a name on what is wrong with it. Since photos are 3D-looking two-dimensional visuals, we are stuck with that angle once recorded on the camera’s sensor. The tools we use enable us to manipulate these images beyond what was thought to be possible a few years ago. However, some properties can’t be changed.

Perspective: In an image, all lines meet at a point. We call this vanishing point, and its part of perspective in an image. If the vanishing point of the two photos to be composited meets at the exact location, then good our brain will take perspective as “correct”; if not, we’ll feel that something is “off.” That means, if possible, there is usually only one place that the two images can be placed -composited. Unlike a 3D object, the object can’t be rotated to change the vanishing point. Therefore as post-productions artists, we are limited where the composite can be placed. Accordingly, we can’t place assets on any background you like; we must make sure they fit into the vanishing point rule.

One of my first acceptable composites, that I matched both vanishing point and light direction matching.

The quality and direction of light: In photo editing tools (Adobe Photoshop, Affinity, or GIMP) direction of light can’t be manipulated. This rule is a bit easier to understand. If, in the background, the light is coming from the left but the clipped subject image the light is on the right, and if we place them on top of each other again, viewers will immediately understand the picture is fake. The remedy is sometimes flipping the image horizontally, but that may not be possible on every shot. The photographer should pay attention to light quality and direction.

Other than these two critical, almost impossible to manipulate issues, the rest can be solved in software. The talent should pay attention to details and be on the lookout for how light behaves around objects in real life. I often see composites on the digital media, even on the billboard of some well-known big brands with big ad agencies on their roster, and I can’t believe how those images got approved. 

So what are the common mistakes other than light direction & perspective issues?
One obvious mistake I see is color. When image compositing is involved, environmental light should affect the subjects. Therefore any added item must show this effect. 

Shadows are also a common mistake. First of all, contrary to our perception, shadows are not gray or black. They are darker tones of the surface they fall onto. Therefore they shouldn’t be inserted as black or gray. Also, shadows have layers closer to the subject; they get darker and softer and lighter further away.

Glossy or semi-glossy surfaces should reflect the subject, and if the surface is not smooth, for example, ripples on a wave that way, the reflection should have those effects.

Composite Critics

I checked a couple of composites on the Internet and f here are some of my critiques.

Istanbul Municipality Women Firefighters image: Good sky replacement but sky colors have not been implied on the foreground, colors are cooler on the subjects.

Istanbul Municipality Women Firefighters. Good sky replacement but we can’t see the reflection of the sky on the windows of the truck.

BMW ad. Front trees have motion blur but the trees behind the car in the middle is static.

Car has a shadow but where is the shadow of the driver?

My Composites

As I have criticized other works I should share some of my photo compositing work and I’m sure there are mistakes I have missed. Maybe you can catch them.