Novel view synthesis in embedded virtual reality devices

Virtual Reality and Free Viewpoint navigation require high-quality rendered images to be realistic. Current hardware assisted raytracing methods cannot reach the expected quality in real-time are also limited by 3D mesh quality. An alternative is Depth Image Based Rendering (DIBR) where input only consists of their associated depth maps for synthesizing virtual views Head Mounted Display (HMD). The MPEG Immersive Video (MIV) standard uses such DIBR algorithm called Reference View Synthesizer (RVS). We have first implemented a GPU version, Realtime accelerated (RaViS), that synthesizes two HMD. In present paper, we explore differences between desktop embedded platforms, porting RaViS an HMD without need separate, discrete GPU. proposed solution gives insight into Synthesis techniques HMDs using OpenGL Vulkan, cross-platform rendering library with support devices.