This repository contains projects from the CS 523 - Computer Vision couse in Ozyegin University.
JPEG compression to four given grayscale images and observation on the changes in frequency domain.
In this assignment, goal is classifying the action in a given video. In this link (http://www.wisdom.weizmann.ac.il/~vision/SpaceTimeActions.html) you can download the dataset you will be working on. Note that you need to download all of the classes in the Classification Database. On this dataset, you will be applying HOOF [1] (Histogram of Oriented Optical Flow), PCA (Principal Component Analysis) and k-Nearest Neighbors (k-NN) for classification.
[1] Chaudhry, Rizwan & Ravichandran, Avinash & Hager, Gregory & Vidal, René. (2009). Histograms of oriented optical flow and Binet-Cauchy kernels on nonlinear dynamical systems for the recognition of human actions. 2012 IEEE Conference on Computer Vision and Pattern Recognition. 1932-1939. 10.1109/CVPRW.2009.5206821.