Fourier Transforms are amazing and fun! I want to detect if a player is present in VR but not moving very much. It turns out to be _very hard_ to pull apart small player movements from tracking noise. After many attempts I remembered vaguely that Fourier transforms might do this
-
Show this thread
-
So I went and found this lib: https://www.codeproject.com/articles/1107480/dsplib-fft-dft-fourier-transform-library-for-net … I fed some player position into the example code and it "kinda" worked. After some tweaking it works to a level that is near magical. Tweaks: I feed it the rotation quaternion's W value as it's more general than any one axis
1 reply 0 retweets 7 likesShow this thread -
I discard the first result as it represents the magnitude rather than the fluctuations I am looking for. I take a sample every frame for 120 frames which seems to capture most human oscillations. I just compare the average value of the first 20 results to some value (.0005)
1 reply 0 retweets 2 likesShow this thread -
It's _unclanny_ how well this works. I've tried it with a bunch of different headsets. It's hard to trick it on purpose. I'm sure I sound like an idiot to anyone who knows anything about this but having not played with them ever I was surprised how simple and effective they are.
1 reply 0 retweets 5 likesShow this thread
If you check out Parkinson's tremor measurement apps they give you a full frequency spectrum of shakes. Was interested to know frequency also matters. (But didn't look more into it)
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.