9 frames are shown below. The depth-of-field is varied from 0 to 1 (0 being furthest and 1 being closest). This sequence is created by shifting each image by the calibration data scaled by the depth-of-field factor. The first image is held as reference to calculate the appropriate shifts for each corresponding image.
The resulting sequence stringed together in a gif
9 more frames are shown below. This time, the aperature was varied from 0 to 1 (0 being smallest and 1 being largest). The sequence is generated by taking a fraction, calculated using the aperature ratio, of the images and taking the average to produce a sequence with the same focus point. The fraction starts as approximately 10%, and increase by 10% until all the images are included in the aggregation.
The resulting sequence stringed together in a gif
It's amazing how such simple calculations and theory can produce incredible results. As a photographer, I found the aperature adjustment effect particularly interesting. The resulting image is similar to the image produced by a tilt shift lens. Now I won't have to spend lots of money on expensive equipment; I can just reproduce the results using code :-)