Most implementations of 3d gaussian splats are static. They are based on pointclouds and not polygons. As these are captured with images and generated from them, the process has no semantic understanding of its content.
There is no technical way to rig each flower or move vertices like in traditional 3d animation.
It is mainly a pointcloud with no segmentation.
But there are projects working on the semantic part, which could open a way to animate the detected objects individually in future.
I read a lot of comments talking about „getting down the operational costs“ but i am missing someone talking about the costs of depositing the nuclear waste until it has no more risks. Am i missing something?!
if the resolution per eye is 4k, that would mean that you need to scale the screen to the whole fov of the hmd and stop moving, because that might blur the content on the virtual screen. i think, apple has put a lot of effort to get around the bluriness but there wont be more pixel available than there are on the little screens. So, its probably a downgrade in resolution for you
they will see an avatar of you with some kind of facial expression and eyetracking, but with a different name (not avatar) of course. But they wont see a videostream of you, what is kind of akward imho.