Video Tracking algorithm using Mean Shift
Steps for execution -
- Select the desired Region of Interest to be tracked throughout the video sequence.
- Use the mouse cursor to draw a bounding box around the object. This code assumes that the bouding box has been drawn by chosing the upper left corner pixel first and then lower right corner.
- Chose the desired kernel properties for density estimation.
- The algorithm will track the selected feature over the frames and sequence them back into a video file.
Please use Mean_Shift_Video_Tracking_OpenCV_2.9.py to be used with version OpenCV 2.9.* versions.