Does it mean they apply some kind of human perception model (like audio codecs apply a psychoacustic one) to determine what detail can be omitted without future viewers potentially noticing the difference?
> zipping a RAW frame would not be possible
why? I mean it's just a big blob, if there's a lot of similar substrings in it, it might give a few percent compression ... also, nonsense doesn't mean non-patentable, right?
> Does it mean they apply some kind of human perception model (like audio codecs apply a psychoacustic one) to determine what detail can be omitted without future viewers potentially noticing the difference?
I have not looked into it too deeply but it appears to be based on wavelet compression (more or less a copy of JPEG2000). They are able to achieve much better compression ratios. I am restricted to lossless compression (and in real time on a mobile device).
> why? I mean it's just a big blob, if there's a lot of similar substrings in it, it might give a few percent compression ... also, nonsense doesn't mean non-patentable, right?
What I mean is that it is unlikely that any form of compression of RAW video data is encompassed by their patent. But who knows.
Perhaps it could also be argued their patent covers cameras and their manufacturers, not 3rd party software that users can install on their phones? Also don’t think MotionCam has enough users for their lawyers to care. Either way thank you for your software, it’s dope
Does it mean they apply some kind of human perception model (like audio codecs apply a psychoacustic one) to determine what detail can be omitted without future viewers potentially noticing the difference?
> zipping a RAW frame would not be possible
why? I mean it's just a big blob, if there's a lot of similar substrings in it, it might give a few percent compression ... also, nonsense doesn't mean non-patentable, right?