I recently got a OpenBCI Cyton board, and I’ve been experimenting with predictive cursor control and keyboard input using GRU networks.
Right now the accuracy is kinda shit, mostly because I haven’t gathered enough training data yet. I don’t expect it to reach the same performance level as traditional tablet or mouse play, but I’m really curious how osu! might handle BCI's in the future.
Here’s what I’m currently doing:
Targeting the occipital lobe, prefrontal cortex, and motor cortex
I am trying to generate usable signals without relying on mental imagery or visualizations (which are slow and tiring to sustain)
I am using abstracting cursor movement into 8 axis (this is so it even possible to get decent accuracy reading with non-invasive BCI)
Current system latency is around 35 ms, which is slightly noticeable but tbh I might have to increase latency even more for better accuracy.
So, I have two main questions for discussion:
1. do you think BCI (brain computer interfaces) should be allowed for ranked play in osu!
2. Would you use a BCI
Right now the accuracy is kinda shit, mostly because I haven’t gathered enough training data yet. I don’t expect it to reach the same performance level as traditional tablet or mouse play, but I’m really curious how osu! might handle BCI's in the future.
Here’s what I’m currently doing:
Targeting the occipital lobe, prefrontal cortex, and motor cortex
I am trying to generate usable signals without relying on mental imagery or visualizations (which are slow and tiring to sustain)
I am using abstracting cursor movement into 8 axis (this is so it even possible to get decent accuracy reading with non-invasive BCI)
Current system latency is around 35 ms, which is slightly noticeable but tbh I might have to increase latency even more for better accuracy.
So, I have two main questions for discussion:
1. do you think BCI (brain computer interfaces) should be allowed for ranked play in osu!
2. Would you use a BCI