These are projects I did for a Computer Vision course.
The goal of this project is to create an image classifier that given an image, can decide which class it fits into.
The provided starter code is in starter_code. My code is in image-classifier/code. I wrote everything in utils.py other than load_data, and I wrote the code to run KNN and SVC on the images in main.py.
Results contains .npy files that contains the descriptor centroids produced by KMeans and Agglomerative Hierarchical clustering, histograms created using a bag of words approach on image descriptors, and the accuracies and runtimes of my classifiers.
Install the version of OpenCV that still has SIFT and SURF
$ pip3 uninstall opencv-python
$ pip3 install opencv-contrib-python==3.4.2.16
The goal of this project is to create a bokeh filter effect by synthetically enlarging the aperture of a camera by taking a video.
There was no provided starter code. Results can be found in results.pdf which is a modified version of spec.pdf.