In this project, we tackle the image compression problem using K-Means Clustering. It is a clustering algorithm that clusters the given data points to k clusters. K-Means's aim is to optimize the position of the cluster centroids, which minimizes the cost function used e.g. sum of squared distances. K-Means works in an iterative way. First, it starts with randomly assigned cluster centroids and iteratively optimizes the position of the centroids. K-means stops when it converges or when the maximum number of iterations is reached. Besides, we interpret our results and use different metrics to choose the best possible value of K.