Depth camera sensor

[tthil]

In the example of the Classification Data :
"//Here you would grab some data from your sensor and label it with the corresponding gesture it belongs to""

If I would like a sample a left swipe gesture what data would I be looking for in my sensor video stream?

Would the closest point (x, y, z) be a good candidate to sample or do I need more data per frame?

//Thomas

[nickgillian]

Hi Thomas,

The closet point (x, y, z) over time might be a good option, but this will depend on how noisy this signal is and how complex your gestures are.