One of the big problems that the computer vision field still faces today is stereo. Stereo is having two different focals points looking at the same things from two slightly different angles to merge into one focal image. The most common example is our eyes. The left and the right eyes see slights different pictures but our brain can merge them flawlessly. Even the most recent methods are flawed in some parts and not fast enough to have any commercial use. This was my class's effort to optimize a current stereo solution.
"Homework2.Streo.pdf" documentation shows the steps taken to find the disparity, distance between two focal points.