Page status (reviewing guidelines)
This page is a copy of the same page in 2.4 manual, need to be updated
Compositing Nodes allow you assemble and enhance an image (or movie) at the same time. Using composition nodes, you can glue two pieces of footage together and colorize the whole sequence all at once. You can enhance the colors of a single image or an entire movie clip in a static manner or in a dynamic way that changes over time (as the clip progresses). In this way, you use composition nodes to both assemble video clips together, and enhance them.
We use the term Image to refer to a single picture, a picture in a numbered sequence of images, or a frame of a movieclip. A node layout will process a sequence one image at a time no matter what kind of input you provide.
To process your image, you will use nodes to import the image into Blender, change it, merge it with other images, and finally save it.
The example to the right shows the simplest noodle; an input node threads the camera view to an output node so it can be saved.
"Nodes" are individual blocks that perform a certain operation, and might have one or many different outputs.
Conceptually, there are three basic types of nodes:
- Input Nodes
- these nodes produce information, but do not have any inputs of their own.
- Examples are: Render Layers, Value and RGB nodes.
- Processing Nodes:
- these nodes filter or transform their inputs, to produce one or more outputs.
- Examples are: RGB Curves, Defocus, and Vector Blur nodes.
- Output Nodes:
- these nodes consume their inputs to produce some kind of meaningful result.
- Examples are: Composite node (which determines the final output used by Blender), Viewer (which displays the output of a socket), and File Output node.
The essential idea of nodes is that you can create an arbitrarily-complex network of nodes, by connecting the outputs of one or more nodes to the inputs of one or more other nodes. Then, you can set appropriate parameters (as you see fit) for each node. This network is called a "noodle" and it describes how information literally flows through to produce whatever result you want.
You can define node groups, and use those groups as they were a single node. You can link and append these node groups from other files.
Accessing and Activating Nodes
Access the Node Editor and enable Composite Nodes by clicking on the Face icon.
To activate nodes for the composition, click the Use Nodes button. Blender creates a default starting noodle, consisting of two nodes threaded together.
To use this mini-map, you must now tell Blender to use the Compositing Node map that has been created, and to composite the image using composition nodes. To do so, switch to the Scene button area (F10) and activate the Do Composite button located below the Animation button. This tells Blender to composite the final image by running it through the composition node map.
You now have your first noodle, a RenderLayer input node threaded to an Composite output node. From here, you can add and connect many types of compositing nodes, in a sort of map layout to your heart's content (or physical memory constraints, whichever comes first).
You can do just about anything with images using nodes.
Raw footage from a foreground actor in front of a blue screen, or a rendered object doing something can be layered on top of a background. Composite both together, and you have composited footage.
You can change the mood of an image:
- To make an image 'feel' colder, a blue tinge is added to the pic.
- To convey a flashback or memory, the image may be softened.
- To convey hatred and frustration, add a red tinge or enhance the red. The film 'Sin City' is the most extreme example of this I have ever seen.
- A startling event may be sharpened and contrast enhanced.
- A happy feeling - you guessed it - yellow (equal parts red and green no blue) for bright and sunny.
- Dust and airborne dirt is often added as a cloud texture over the image to give a little more realism.