For the layman – How 3D Rendering Works. Part 5: Camera

[ about a 15 min. read ]
Part 1: 3D models
Part 2: Materials & Shaders
Part 3: Textures
Part 4: Projection mapping & UV mapping

In the previous article I talked about UV mapping. what it exactly is and how it influences texturing. In this article however we are going to address a whole different topic. We are going to talk about the camera.

The camera is very intertwined with composition, but to keep the articles short, camera and composition will be split up in 2 parts. This part will focus on the camera. Specifically the aspects that are of importance for controlling the look of a 3D scene. The next part will cover composition. Just be aware that camera and composition overlap a lot.

1. Camera lenses/Focal length

In a 3D render application, one of the first things to choose when you want to start your render, is the Focal length of your camera lens.
In short: Focal length is dictated in millimeters (mm). It indicates the field of view you have (how much of the scene will be in view) and what the magnification factor will be (how large the elements in the scene will be). The rule is: the higher the focal length in mm the narrower the field of view and the higher the magnification.

Lenses are normally categorized in 5 different groups of focal length. Each used for a specific application:

  • Ultra wide angle lens [10-24 mm]: Rarely used. In real life sometimes used for interior photography for small rooms.
  • Wide angle lens [24-35 mm]: Landscape and architecture photography.
  • Standard lens [35-70 mm]: Used for portraits, characters, standard objects and vehicles. 50 mm is the focal length of a human eye.
  • Telephoto lenses [70-300 mm]: Used for macro shots and large vehicles. In real life often used for wildlife photography.
  • Super telephoto lenses [300> mm]: For massive vessels and aerial shots from a large distance. In real life also often used for small distant objects (e.g. small birds).
  • Below you see an example for 4 different focal lengths for a car render:

    24 mm

    35 mm

    70 mm

    200 mm

    As you can see using different focal lengths gives a completely different feel to the shot. It changes the perspective in the shot, because the field of view and magnification changes with the different focal lengths. It is therefor a bit of a balancing act between choosing the right focal length and moving the camera back and forth to get the perfect camera position. For example for the 24 mm focal length the camera was moved way closer to keep the same amount of the car in view. For the 200 mm focal length the camera was moved way back to keep the car the same size in the image.

    For most of my visualizations I use a focal length between 35-70 mm. Most of the time I use the 50 mm focal length. This gives a very natural look as it is the focal length of the human eye.

    2. Camera position

    The next thing you can control with your camera is the position. The position, together with the focal length of the camera, is strongly going to influence how the scene is perceived. But the positioning of your camera depends a lot on the subject of your shot.

    For life size familiar objects and environments, like architecture interiors and exteriors, or vehicles and items placed on a table, you might want to consider positioning the camera at a familiar standing or sitting eye level height. These averages are of course different all over the world, but for standing you can use a value between 150-170 cm. And for sitting between 74-82 cm.

    The previous car renders were taken from a low camera angle around 40 cm from the ground. Now we can put the camera at an eye height of 170 cm, to give a more familiar perspective on a car:

    Camera at ground level: 40 cm

    Camera at eye level: 170 cm

    Now for a car the lower camera angle gave a more artistic feel to the image. It is a position often used for car visualizations. But if we would use the same for an interior visualization you will notice that it doesn’t really work that well. An eye level perspective gives a more natural view:

    Camera at ground level: 40 cm

    Camera at eye level: 170 cm

    The low camera angle really gives you the feeling like you are lying on your belly on the floor.

    For a product shot things can be drastically different again. You can have a more close-up shot from a seated position:

    Japanese Tea Set High Front View

    Or a more dead center low camera position, for a bit more of a dramatic effect:

    Japanese Tea Set Front View

    The most important point I want to convey here is, that the position of the camera is very relative to the shot you want to make and the product you are trying to display. There are some practical guidelines as you saw in the examples before that work better for one product than for the other. But try to do what feels natural to you and often or not you are already well on your way to a good camera position.

    By the way, this is already very much part of the next topic on composition. As I said before, camera and composition have a huge overlap.

    3. Depth Of Field (DOF)

    Depth Of Field, or abbreviated as DOF, is a powerful technique that can add realism to your renders or create specific focus on a part of your render. It is basically creating focus on a specific point in your scene and blurring the foreground and background around it.

    In real cameras the DOF depends on the aperture size of the lens (roughly speaking the size of the opening of the lens, and it is also know as f-stop). By changing the aperture you would need to control the change in light input through the lens to control the exposure of the final image (I would recommend this article if you want to know more about camera DOF). Now In 3D we do not need to worry about any of this, as DOF is simply controlled by a slider. Very simple, and a huge benefit of working in 3D.

    The following 2 examples show you how adding DOF creates emphasis on the point in your scene where you want your audience’s eye to focus on.
    In the first comparison image below you can see a production skid with a lot of tubing. The image comes across as very cluttered due to the proximity of all the tubing. Now the goal of the image is to actually focus on the tubing at the front. By adding DOF we can put the focus exactly on that. By adding the DOF the image not only serves its goal, but it also becomes more gentle on the eyes.

    The same can be said for the next example of a circuit board component. The goal is to put emphasize on the red button:

    That’s it on cameras. As always, I hope you enjoyed this post and hopefully you have learned something new. Let me know if you have any questions or comments below.

    Leave a Reply

    Your email address will not be published.