Have you noticed VR video just doesn’t look very good?
There’s nothing wrong with your eyes. VR video on the latest headsets looks pretty bad. Even video captured at 4k doesn’t look sharp. That is because the video has to be stretched and scaled onto the bigger 360-degree canvas the headset affords. To get a full stereoscopic experience the resolution gets cut in half and sent to each eye and gets stretched again vertically. By the time all of this scaling and distortion occurs the visual quality is severely compromised. VR video looks worse than the early days of online video – not even Standard Definition. Headset displays will soon be 4K and VR content will continue to improve with original resolutions soon exceeding 8K and more! Your streaming solution needs the power to support these huge bit rates to deliver a truly immersive VR experience. The required files sizes and bit rates are massive and will continue to grow over time. And these files will be impossible to stream in real time.
Close up Magic
Rescaling and compression alone will never solve this problem. The file sizes will still be huge. Most VR streaming schemes attack this problem by sending only what the user sees in the field of view on the headset at the highest quality, and processing what is not visible at lower quality. When the viewer changes the field of view with head movement, the new field of view is displayed at the higher quality. There are lots of ways to slice and dice this. But very few deliver true stereoscopic Virtual Reality presence in real time, on the fly. After optimizing every frame at the macro block level, every Full VR stream is uniquely personalized, taking into account network conditions as well as what the viewer is seeing and when they are seeing it. This is all accomplished by harnessing the computational power of Cinova’s cloud server technology, resulting in minimal latency or visual lag and true VR presence.