In this project, we implement tiny images and bag of words as our feature extractors, and nearest neighbors and SVM as our classifiers. Using bag of words to extract the features and SVM classifier to classify the images yielded an accuracy of 65.265% which was the highest amongst all other combinations.
