PODDEM is a state of the art detection and estimation method for outliers in Particle Image Velocimetry data. PODDEM can be used on two-dimension and three dimensional data, with up to three velocity components.
The details of the algorithm are published in:

Higham J.E., Brevis, W., Keylock, C.J. A rapid filtering and reconstruction method of two-dimensional image velocimetry signals using a non-iterative POD-method. Accepted, Measurement Science and Technology (IF: 1.43).