Assignment 3 - Vision, Cameras & Color (100 pts)

Due March 22 by email to

1a.) Why is computer vision a difficult problem?
b.) Explain how we might use clustering to segment an image, including one possible similarity measure you might use.
c.) What are some applications of computer vision that work well? What applications do not?

2a.) Why in a pinhole camera does the image appear upside down?
b.) Why don't we see everything around us as upside down?

3a.) What does perspective projection mean?
b.) What ambiguities does it cause?
c.) Why when we're drawing a perspective projection do we usually draw the image plane in front of the center of projection?
d.) What are homogeneous coordinates and why do we use them?

4a.) Why do pinhole cameras produce blurry images?
b.) What is an aperture?
c.) Why might you not want to make the aperture as small as possible?
d.) What can you add to a pinhole camera to help reduce this blurryness?

5a.) What does focal length refer to?
b.) What does circle of confusion mean?
c.) For a camera with focal length of 200mm, if an object is located at a distance of 400mm from the lens, and the film is located 200mm from the focal point, will that object be in focus? How about if we move the film 200mm further away?
d.) Why do faces look egg-shaped in cell phone cameras?

6a.) What does this statement refer to: "At night, all cats are gray!"?
b.) Explain the differences between rods and cones.
c.) Explain why when you look up at the night sky there appear to be more stars off-center?