Multi-view rendering is the process of generating all the perspective views present in the projection of a scene regardless of viewer position. This can require 1,000s of views to be rendered per update for a light-field display and binds the display to a particular application.
Object Graphics Language (ObjGL), in development at FoVI3D, is an application and display agnostic API where rendering becomes the responsibility of the display. ObjGL draws heavily from the popular OpenGL graphics language yet is streamlined and optimized for fast rendering for remote multi-view systems. The ObjGL API is simple, efficient, and provides many geometric rendering clues that can be exploited by a well implemented multi-view rendering system.
Decoupling the host application from the display enables the heterogeneous display environment (HDE). Within the HDE, the host application executes blindly without knowledge of the number or architecture of attached displays. This decoupling of rendering responsibilities from the host allows for the greater selection of display technology for a given application. As the word “heterogeneous” implies, the developed solution is applicable to all forms of novel 3D display technologies, including FoVI3D’s LfDs, Oculus Rift HMD, LightSpace’s multi-planar volumetric display, etc.