How do we understand the image?

Are we allowed to modify the input image data? Please  look at + for 1 min, you will see magic

Are we using priori knowledge model? 

Are we using our experience? 

Are we using some comparison algorithm? (Point B and A is the same color!)

Do we make conclusion based on continuity? (see big illusion images)

Is it based on logical experience?

What is actually a movement?

bg subtraction

Simple Bg subtraction
in cvTracker

First Compressed Sensing Reconstruction

(under construction)

(Brift Theory)
(Parameter Details)
(Tool Box)
(Method)
Result

Discussion
          The reconstructed signal seems to be changing every run. The correlation between reconstructed signal and the original signal seems to be very high except for the edge step peak, which seems to get high error almost all the time.