CS143 Introduction to Computer Vision Project 1 - Hybrid Images with laplacian pyramids Hung-I Chuang Login:hichuang

CS143 Introduction to Computer Vision Project 1 - Hybrid Images with laplacian pyramids

Hung-I Chuang Login:hichuang

Introduction

Hybrid images are static images that change in interpretation as a function of the viewing distance. The basic idea is that high frequency tends to dominate perception when it is available, but, at a distance, only the low frequency (smooth) part of the signal can be seen. By blending the high frequency portion of one image with the low-frequency portion of another, you get a hybrid image that leads to different interpretations at different distances.

Method

First of all, we need to find a couple pairs of images that we want to hybrid them together. We then have to generate Gaussian and Laplacian pyramids for both images. After we get two Laplacian pyramids, we will need to find a cutoff-frequency to choose levels of low-pass filtered and high-pass filtered to make a best hybrid result.

To generate Laplacian pyramids, we first have to generate Gaussian pyramids. I blur an image using Gaussian filter and subsample the filtered image by half size to make the next level image. Repeat the step until you get enough levels of Gaussian pyramid.

Laplacian pyramids are high frequency of each level of Gaussian pyramids that are filtered. In order to get the high frequency image, we need to subtract two levels of Gaussian pyramid to get high frequency signal.