Understanding VR post-production
/After your VR shoot has taken place, there’s often a bit of a gap before you see the finished product. We all know shooting in VR is really tricky … but what goes on post-production and why does it take so long?
Trying to understand how VR films are actually created is really valuable when thinking of commissioning a film and concept planning. Don’t know the difference between ‘cloning’ and ‘masking’? Below is your cheat sheet to the main pre-production tasks…
Syncing & Stitching
360 video is achieved by using either multiple cameras or a single unit comprising of multiple lenses. The footage is then synced (if using multiple cameras) and then stitched together. Stitching can be a really lengthy process and is not required in conventional 2D filming. The scene and camera selection can have a huge impact on the complexity and difficulty of the stitch. The more complicated the scene, then more time-consuming and expensive the post-production process. For example, tighter environments and scenes where there are people walking around the camera would be considered complicated.
Rendering 1st pass
Once the best stitch possible has been achieved, the footage needs to be rendered into an Equirectangular format ready to begin post-production. It's often necessary to render out separate passes from each individual lens/camera for use in compositing. Depending on the specification of the computer and the length and resolution of the footage, this can be a lengthy process.
Compositing
Stitching software will only get you so far. In certain situations, the stitch may need to be improved manually by covering the stitch lines that the stitching software wasn't able to correct. These are known as 'stitch errors'. This can be done is several ways, depending on the type of stitch to be remedied. These are:
Cloning - duplicating parts of the scene to cover stitch errors
Masking - using digital masking techniques to cover or hide stitch errors
Mesh Warping - using a mesh based grid of Bezier patches allowing manipulation of the footage in specific areas.
Audio Syncing
Because the audio recording through the cameras is often of bad quality, a sound recordist will often be used at the shoot. During the post-production process, the editor must accurately sync the audio provided by the recordist to the footage.
Grading & Colour Correction
Together known as 'grading', this is where the colours are manipulated to enhance the overall visual aesthetic of the film.
Rendering 2nd pass
This is generally a longer process to 1st pass rendering depending on how many applications are used within the edit. Rendering this final stage is often done overnight, especially when the budget and timescales are limited.