sixFaces: The pictures of six faces (3 men and 3 women). To be used to...

In HerveAbdi/data4PCCAR: Some data sets and R-functions for PCA and CA to accompany Abdi & Beaton (to appear, 2023) Principal Component and Correspondence Analysis with R

The pictures of six faces (3 men and 3 women). To be used to illustrate PCA or PLS on images.

Description

sixFaces: The 230*240 = 55,200 pixels pictures of six faces (3 men and 3 women). To be used to illustrate PCA or PLS on images.

Usage

data("sixFaces")

Format

A list containing 3 objects:

• face.array.raw a 230 * 240 * 6 array storing the 6 (230 by 240 pixels) raw images of the faces. Values of the pixels go from 0 to 253.

• face.array.normed a 230 * 240 * 6 array storing the 6 (230 by 240 pixels) normalized images of the faces. The sum of the squared values of a normalized image is now equal to 1.

• face.matrix a matrix of dimensions 6 by 55,220 ( = 230*240) storing the vectorized (normalized) pictures of the faces.

Details

sixFaces contains the original pictures (as a 230 * 240 * 6 array), the normalized pictures of the faces (such that the sum of squares of the pixels of a face is equal to 1), the 6*55,200 matrix of the faces.

Note

To get back a 230*240 image-matrix from the vectorized version of the face use, for example:

face1 <- matrix(face.matrix[1,],230,240,byrow = FALSE)

To visualize an image use, for example:

image(sixFaces$face.array.normed[,,1])

To fold back an image from the vectorized and vizualize it, use:

image(matrix(sixFaces$face.matrix[1,],230,240,byrow = FALSE))

Author(s)

Hervé Abdi

References

The faces have been used in a few publications including:

Abdi, H., Valentin, D., O'Toole, A.J., & Edelman, B. (2005). DISTATIS: The analysis of multiple distance matrices. Proceedings of the IEEE Computer Society: International Conference on Computer Vision and Pattern Recognition. (San Diego, CA, USA). pp. 42-47.

Abdi, H., & Valentin, D. (2006). Mathématiques pour les Sciences Cognitives, (Mathematics for Cognitive Sciences). Grenoble (France): Presses Universitaires de Grenoble.

HerveAbdi/data4PCCAR documentation built on Sept. 11, 2022, 4:19 p.m.