3D compositing (very important process to create a 3D configurator) is a process in which various elements such as live action footage, photographs or CGIs are combined to form a single image or scene. This type of compositing is often used to create still images that combine different components into a single image, or in post-processing for film and television programs to combine different elements into a video sequence. 3D compositing differs from 2D compositing in the way the different layers interact and influence each other.
The following tutorial illustrates the 3D compositing of a 3D animation:
3D composition is usually performed using 3D software developed specifically for such compositing. The input usually comes from different sources. The compositing process typically involves multiple input files, including still images and video files that are assembled and overlaid during the 3D compositing process. A simple example of this process would be an image visualizing a boat on the water in front of a large cliff with with clouds and sky in the background. The finished image shows all elements in a single, seamless image.
Compared to 2D compositing, 3D compositing allows the individual layers from different sources to be joined together in such a way that the elements overlap and interact more realistically. In the example above, the boat could cast a shadow on the water below it. Such visualizations are not possible with 2D compositing.
With 3D compositing, these individual layers can be joined together to create a shadow for the boat on the water. This is done either by a 3D software that can create a shadow based on the boat or by creating the boat in a 3D program, because the generation of a shadow in this program can also be integrated into the scene. Other effects can also be more easily applied to a scene or still image by 3D compositing, e. g. when objects in one layer affect the illumination in another layer. This type of compositing is often used to create more realistic images with 3D software without overtaxing rendering computers by rendering multiple passes for a scene or object, which are then assembled together to form a finished image.