Homework#1: practice an image segmentation algorithm using K-means

Requirement:

1. Use a relative path for reading the color face image which can be downloaded from the Coursework.

2. Use k-means to segment faces from an image, e.s, set k=3 or try other k, note: the second testing image, i.e., faces.jpg is optional which is a little bit challenging, but you will get an extra credit.

3. Find the face cluster and convert the original image to a binary image(face and background).

4. Find the location of face from the binary image.

5. Output of the script should be an image with a bounding box

1. Matlab named UNI_Namekmeans.m/.py (see the pseudoCode.m in the coursework), which must be runnable (we won't debug for you).

2. A readme file (UNI_Name_kmeans.README) Explain the logic behind the Flow (or script), e.g. what is the loop for? What is that piece of code for? (you can do this in the Matlab or Python script directly) Specify and Explain ALL key variables/parameters used in the code. (you can do this in the Matlab or Python script directly) Briefly discuss the limitation(s) of the current script. How can you improve it