Final Project

Avni Prasad, CS194-26-aej

Neural Style Transfer

For this project, we re-implement Gatys et. al.'s paper: A Neural Algorithm of Artistic Style, which uses feature maps in CNN to change the style of a picture, but retains the contents of the original image. We are given two images as inputs and output a new image that reflects the contents of one but the artistic "style" of the other. Go to Project

Content Image

Style Image

Stylized Content Image

Light Field Photography

In this project, we harness simple techniques of depth refocusing and aperture adjustments described in Ng et al.'s paper: Light Field Photography with a Hand-held Plenoptic Camera to create the effect of light field cameras. Light field camera's capture 4D light fields instead of 2D flat images, but we can repoduce this effect using simple operations on 2D images with varying focus length and aperture size taken from different positions on a plan orthogonal to the optical axis. Go to Project

Depth Refocusing

Aperture Adjustment