Section: New Results

Noisy input filtering for interactive systems

Participants : Géry Casiez [correspondant] , Nicolas Roussel.

Noisy signals occur when an original time varying value undergoes undesirable and unpredictable perturbations. These may be caused by things like heat and magnetic fields affecting hardware circuitry, the limits of sensor resolution, or even unstable numerical computation. Noisy signals are a common problem when tracking human motion, particularly with custom sensing hardware and inexpensive input devices like the Kinect or Wiimote.

We developed the 1€ filter ("one Euro filter") is a simple algorithm to filter noisy signals for high precision and responsiveness. It uses a first order low-pass filter with an adaptive cutoff frequency: at low speeds, a low cutoff stabilizes the signal by reducing jitter, but as speed increases, the cutoff is increased to reduce lag. The algorithm is easy to implement, uses very few resources, and with two easily understood parameters, it is easy to tune. When compared with other filters, the 1€ filter shows less lag for a reference amount of jitter reduction [15] .

The 1€ filter is already used on a daily basis by many other researchers and companies.