RGB Color Space

The purpose of this assignment is to get you comfortable thinking in the RGB color space. In class we made some histograms of the different RGB colors present in the image of Stitch. Pick 3 of your own images (from the web, from your own photos) and create a visualization that shows how similar/different these images are using these histogramming methods presented in class.

Note: it is possible that one or more of the images you use may not have an “alpha” channel and will only have RGB triplets. You will then have to modify the code we used in class and take out any references to the “alpha” mask.

You are free to choose 3 images that look “similar” to you or very different and describe how mapping them into color space (i.e. making histograms of their color distributions like we did for Stitch in class) either confirms or negates these differences or similarities. (For example, let’s say you choose a painting of a beach and a photo of a beach their color spaces might look very different although they are representations of similar objects). You may also want to choose a set of images that you like and those you don’t like to see if there are any differences due to color choices.

The attached notebook & example image are an example of how to downsample your image into its most frequent colors or a user-selected set of colors. While you do not have to necessarily use this for your visualization, you should nevertheless take some time to look over the code and answer the following prompt in a few sentences (3-5 sentences is fine):

Both of the color re-mappings presented in this notebook are not perfect. What are some issues you can think of in how color is rebinned from a full color image into an image with less colors with the functions provided?

The coding portion is worth 15 points, writeup 15 points of your code. Your exploration of the rebinning code is worth 10 points.