I did the following 2 projects: Vertigo Shot, and Fake Miniatures.

In order to produce this part, I essentially had to take a picture of the subject at multiple distances. The farther I was, the more zoomed in I had to take the picture. When taking the pictures, I aimed to make sure the subject had the same pixel dimensions in each shot and in the same position. For example, I would want the subject to be in the middle 300x300 pixel square of every image. I took the shots with an iPhone 7+ camera using ONLY the optical zoom, because the digital zoom created too much grain. This means for each example, I took pictures at 2x, 1.7x, 1.4x, 1x. I started with most zoomed in first. Here are the results.

For this project, I wanted to create "fake miniature" objects in an image. For this, I had the user input a focus line that would be used to focus on the subject that they want to miniaturize. After obtaining the focus line, I created a small rectangle using the line as the bisection. This created the depth of field (DOF), essentially just turning the 1D line input into 2D plane. From there on, I iterated through the portions outside of the DOF in such a way that the farther the pixels were from the DOF, the more blurred they would get. This helps to make the mini effect. As a bell and whistle, I also sharpened the image with the code from project 3 to make the subject more focused. I noticed subjects in tilt shifts had a specific sharpness to them, so I thought it would help. Here are results:

It was really cool to attempt the vertigo shot after seeing it in lecture in the beginning. It was a nice opportunity to get a better understanding of how the effect works. I also enjoyed very much making the fake minis, and although the code is quite simple it works really well for certain pictures with a linear DOF.