Photographers have been “fixing in post” for years but it is still not uncommon for new and old 3D users alike to be using rendered frames as a final output. Here we discuss how a compositing application can save you time, improve efficiency and produce amazing effects.
From a photography point of view, it’s very rare to find any photographer taking raw images and then using these files without any post processing. As such, the standard workflow would consist of photographing the subject, taking images into a photo-editing package (like Photoshop or Aperture) then making adjustments to things like tone, saturation, colour or composition, to ensure the perfect image. In the same way, a good 3D compositing application will ensure that the final output of any rendered frames is as intended. As good as our 3D applications are, it’s still very rare for rendered footage to be perfect straight out of your authoring application. Despite this, many 3D users fail to use even the simplest of post- processing programmes.
A good compositor, such as Adobe’s After Effects, is an essential part of any CG workflow. Many 3D applications allow for the export of rendered out frames into the varying compositing applications native file format, making for easy integration and an uninterrupted pipeline.
By altering the way in which scenes are rendered, the true extent and capabilities of a compositor quickly become clear. If rendering out a scene into layers isn’t something that you are already doing, it is a vital step that should be integrated into your workflow. From the early days of 2D hand drawn animations on acetate, working in layers has been a standard practice, and any one that uses Photoshop will appreciated the sort of benefits this type of work flow will bring.
Now, depending on your 3D application, there are varying options here but the general principles are the same: you decide which parts of your scene you want rendered out, and on what layers you want them rendered out to. For example, you could split your render up into layers such as shadows, secular and diffuse, or into dedicated layers for each object in your scene.
When taken into your compositor, these separate layers mean that the effects and touch-ups can be quickly made to individual parts of the scene without the whole scene being affected. For example, after rendering out a sequence of frames it would not be uncommon for rendered shadows to be to harsh, or not harsh for a given environment. Rather than having to alter the lighting settings in your 3D modelling app and do a re-render, just take the rendered files into your compositor and simply adjust the brightness on the shadow layer until the desired effect is achieved! Simple!
No need to wait for a re-render, something that can save hours, especially if we’re talking about 5 minutes of rendered footage for HDV. This opens the possibilities to quick fixes and colour changes to objects, as well as the application of special effects like motion blur, to specific items within your scene! Its complex effects like this that really do save hours – the thought of having to replicate motion blur in any 3D app, even for the most experienced artist is a nightmare! Your client now wants his car in green, not red – a quick click of the Hue/Saturation slider in your compositor and you’re good to go… No re-render, no wasted time!
I hear what you’re saying – it takes to render each layer. Now, here lies the key: rendering out as a multi pass should take no longer than if you were doing one pass! You application will automatically separate each pass or layer as it processes the render at no extra time cost!
And you’re not just limited to adjusting individual layers. Render your scenes in multiple passes and then post processing them into a compositor, instantly provides depth for the addition of other clips or special effects. For example, if you wanted to add an explosion to a non-layered flat video render, any effect would only be applied on top of the video, i.e. in the viewer’s foreground. By splitting the render into multiple passes, you have the option of your main scene background on one layer and a parked vehicle on another. The explosion effect or clip that is then applied, can be added between these two layers so that the car in the foreground hides any of the explosions that are occurring behind it.
Compositing software will also allow for the simple and effective addition of rendered out 3D frames into pre-shoot video footage, for the interaction of actors with 3D animated characters or objects as well as the ability to use green screens for virtual sets and studios. Empowering directors with the tools to create scenes and visuals previously unachievable with traditional methods. Want to quickly add a CG generated logo to the bottom corner of some video? Not a problem… model the logo in 3D, render it out on an individual layer or using an alpha channel, and then simply add the layer over existing footage in your compositing application and voila, you are good to go!