Creating a CG Car for a Commercial, Film Prop, or Video Game

Toyota_RedThe process to create a computer generated image (CGI) involves hardware and software to produce a quality product. A 3D scanner is first used to capture the model, and then the raw data is processed into 3D modeling software such as Autodesk Maya or exported from AutoCAD. The high resolution data from the scan or CAD file is then optimized for CGI.

Let’s say the body of a vehicle is scanned and the scan or CAD data has higher than 2 million polygons. It then needs to be retopologized to a much lower resolution. A max amount of polys for any vehicle should be between 100,000 polys and 200,000 polys, depending on how close to the camera a shot was taken as polygon density varies. Whenever scan data has holes, all geometry needs to be cleaned up and/or retopologized. One tool we use for this (others are available) is ZBrush. When CAD data is used, inner parts that are invisible to the camera need to be deleted to make the file lighter.

Toyota_White_SilverIn the data prep stage we also start to create UV space for the geometry. If materials such as fabrics are hand painted, there should be a greater focus to get an even UV space so no stretching occurs. We use Headus UVLayout for creating the UVs. After the base model is done, we make sure to check the normals and nonmanifold geometry. We then export the data to Maya for polishing the model.

To scan the exterior of a vehicle takes approximately a week or so, based on detail. If the interior also needs to be scanned, add an additional week or more depending on the level of detail and coverage needed.

Toyota Interior

The product is normally delivered with UVs and  textures/shaders, but can be customized to our customers specific need.

Where’s your closest TNG location?

tng vfx locations

We welcome your feedback! Comment on our blog or visit our Facebook page.