Last modified on 6 June 2009, at 17:01

SwisTrack/Components/MotionTemplateParticleDetector

DescriptionEdit

This isn't actually a tracker but a particle detecter. It emits particles when there is motion in the image, calculated by finding gradients on frame-differences of images over multiple timescales. See the OpenCV book for more information. It also associates the particles with a color histogram if it's given a foreground model in mDataStructureImageBinary.

ParametersEdit

Note: I use default values for all parameters. Read the openCV book and documentation for more details - I wrote this component following the section in the openCV book on motion templates.

DiffThresholdEdit

Threshold to use for frame differencing.

MhiDurationEdit

passed to cvUPdateMotionHistory. "Maximal duration of motion track in the same units as timestamp."

SegmentThresholdEdit

passed to cvSegmentMotion. "Segmentation threshold; recommended to be equal to the interval between motion history "steps" or greater. "

MaxTimeDeltaEdit

passed to cvCalcMotionGradient. "The function finds minimum (m(x,y)) and maximum (M(x,y)) MHI values over each pixel (x,y) neihborhood and assumes the gradient is valid only if min(delta1,delta2) <= M(x,y)-m(x,y) <= max(delta1,delta2)."

MinTimeDeltaEdit

passed to cvCalcMotionGradient. "The function finds minimum (m(x,y)) and maximum (M(x,y)) MHI values over each pixel (x,y) neihborhood and assumes the gradient is valid only if min(delta1,delta2) <= M(x,y)-m(x,y) <= max(delta1,delta2)."