CS 194-26 Image Manipulation and Computational Photography

Defining Correspondences

The goal of this part is to annotate points in both of the images in order to have correspondances when performing image morphing. I defined 43 corresponding points between my face and George Clooney's face.

Images of Me and George Clooney

For image annotations, I used total 47 points, including the four corners of the images. They are shown below

Correspondences

Computing the "Mid-way Face"

To compute the midway face, we define transformations from each of the individual images to an average shape. Then, we use the inverse transformations on each point in a triangle in our average triangulation to get a point in each of our images. To get the average color, we use interpolation to get pixel values from each image and take the average of those values.

Original, The Mid-Way Face, Original

The Morph Sequence

For this section I used 45 frames creating the gif below.

Me to Clooney

The "Mean face" of a population

For this part, we want to find the average face of the population. Using an annotated dataset from the FEI dataset, I took the average of all the annotated keypoints and used that as my average shape. Then, I morphed each of the images, without smiles, into the average shape. As the images are not totally aligned well and morphing into the average shape doesnt regard the colors, some faces some out to be al ittle disfigured.

Original and the morph to the average shape

Original and the morph to the average shape

Original and the morph to the average shape

After doing the two steps mentioned above, we go ahead and cross-dissolve all the images together to find the average face. You can see the result below.

The Average Face

After finding the average face, I morph my image to the average face shape and the average face to the shape of my face. You can see some distortions in the morph of my image to the average shape, which is probably because of correspondence points and my selfie. As you can see on the average images it captured some of my facial features, like the nose and chin.

My image, Morph of my image to average shape, vice versa

Because the alignment of the images is off and the faces are not really the same shape because of how I resized images, the faces look rather deformed in shape.

Caricatures

In this step, we extrapolate from the population mean we calculated and morph my image to get those distinctive features. You can see some examples below.

Caricatures, t = -0.8, -0.2

My nose becomes particularly enlarged and because of color differences in the image, my complexion becomes much darker as well.

Bells and Whistles: Morphing myself into average Korean woman

In this section, I morph my face into the face of an average Korean woman. I morph only warping, only cross dissolving, then with warping and cross dissolving

Average Korean face

Warping Only, Cross Dissolving Only, Both