calcOpticalFlowFarneback

calcOpticalFlowFarneback

Computes a dense optical flow using the Gunnar Farneback’s algorithm.

C++: void calcOpticalFlowFarneback(InputArray prev, InputArray next, InputOutputArray flow, double pyr_scale, int levels, int winsize, int iterations, int poly_n, double poly_sigma, int flags)

C: void cvCalcOpticalFlowFarneback(const CvArr* prev, const CvArr* next, CvArr* flow, double pyr_scale, int levels, int winsize, int iterations, int poly_n, double poly_sigma, int flags)

Python: cv2.calcOpticalFlowFarneback(prev, next, flow, pyr_scale, levels, winsize, iterations, poly_n, poly_sigma, flags) → flow

Parameters:

prev – first 8-bit single-channel input image. next – second input image of the same size and the same type as prev. flow – computed flow image that has the same size as prev and type CV_32FC2. pyr_scale – parameter, specifying the image scale (<1) to build pyramids for each image; pyr_scale=0.5 means a classical pyramid, where each next layer is twice smaller than the previous one. levels – number of pyramid layers including the initial image; levels=1 means that no extra layers are created and only the original images are used. winsize – averaging window size; larger values increase the algorithm robustness to image noise and give more chances for fast motion detection, but yield more blurred motion field. iterations – number of iterations the algorithm does at each pyramid level. poly_n – size of the pixel neighborhood used to find polynomial expansion in each pixel; larger values mean that the image will be approximated with smoother surfaces, yielding more robust algorithm and more blurred motion field, typically poly_n =5 or 7. poly_sigma – standard deviation of the Gaussian that is used to smooth derivatives used as a basis for the polynomial expansion; for poly_n=5, you can set poly_sigma=1.1, for poly_n=7, a good value would be poly_sigma=1.5. flags – operation flags that can be a combination of the following: OPTFLOW_USE_INITIAL_FLOW uses the input flow as an initial flow approximation. OPTFLOW_FARNEBACK_GAUSSIAN uses the Gaussian  filter instead of a box filter of the same size for optical flow estimation; usually, this option gives z more accurate flow than with a box filter, at the cost of lower speed; normally, winsize for a Gaussian window should be set to a larger value to achieve the same level of robustness.

The function finds an optical flow for each prev pixel using the [Farneback2003] algorithm so that

Note

• An example using the optical flow algorithm described by Gunnar Farneback can be found at opencv_source_code/samples/cpp/fback.cpp
• (Python) An example using the optical flow algorithm described by Gunnar Farneback can be found at opencv_source_code/samples/python2/opt_flow.py