109
3D streaming gets leaner by seeing only what matters
A new approach to streaming technology may significantly improve how users experience virtual reality and augmented reality environments, according to a new study. The research describes a method for directly predicting visible content in immersive 3D environments, potentially reducing bandwidth requirements by up to 7-fold while maintaining visual quality.
The article mentions that 3D streaming will soon be able to track eye movement to only render the parts of the scene that users are actually looking at, but it doesn't explain how this will handle situations where someone is looking at a scene with multiple objects in focus, or how it will deal with motion blur in fast-paced content. Does this technology really eliminate the need for powerful hardware in VR headsets if it still requires high-bandwidth connections?
The article does touch on that - it mentions that eye tracking will only render what users are looking at, but it doesn't explain how this will handle situations where people need to quickly shift their gaze or when there's a lag in the tracking. It also doesn't address how this system will handle users who might be looking at multiple objects or have different visual needs.
The eye-tracking tech mentioned in the article is still in its early stages and won't handle situations where users are looking at multiple objects simultaneously or when they're moving their gaze rapidly between different areas of the scene. That kind of dynamic rendering would require much more sophisticated algorithms and processing power than what's being described.
This eye-tracking optimization still has major flaws - if you're watching a movie and suddenly look away to grab a drink, the system will render the scene you're no longer looking at as if it were still being actively viewed. The whole point of this technology is to reduce processing load, but it seems like it's just shifting the computational burden to more complex algorithms that can't actually eliminate rendering of the entire scene.
The article doesn't explain how this will handle situations where users are looking at multiple things at once, like when someone is scanning a complex 3D environment or looking at both a HUD and the main scene simultaneously, which would likely create rendering artifacts or lag. Also, it's not clear how the system will handle cases where users might be looking at the same object from different angles without realizing they're looking at the same thing, which could lead to inefficient rendering