Test Images for the Face Detection Task


Suggested Use

The images listed below are provided in aid evaluating systems developed for face detection. They were used in several papers ([1] [2] [3]) to evaluate our face detection system, which detects frontal faces in grayscale images. Wwe recommend that you use these images for testing purposes only. Training systems based on these images decreases the value of this database as an independent test set.

Please give appropriate acknowledgements when you use these test sets! In the lists of files below, you will find references to Test Sets A, B, and C (together known as Test Set 1 in [1] and the Upright Test Set in [2], and the Rotated Test Set. Test Set B was provided by Kah-Kay Sung and Tomaso Poggio at the AI Lab at MIT, and Test Sets A and C were collected here at CMU (by Henry A. Rowley, Shumeet Baluja, and Takeo Kanade).


Data Formats

In evaluating the performance of a system on the test images below, we need to specify which faces the system is expected to detect. Our work concentrated on finding frontal views of faces in scenes, so the list of faces includes mainly those faces looking towards the camera; extreme side views are ignored. The list of faces to detect are in a file with the following format (one line per face to be detected):
filename left-eye right-eye nose left-corner-mouth center-mouth right-corner-mouth
For each feature on a face to be detected, two numbers are given. These numbers are the x and y coordinates (measured from the upper left corner) of the feature in the image. These coordinates may be used to determine whether each face has been located accurately. The images themselves are all in Compuserve GIF format, and are grayscale. A tar file containing all the images is also available. This file will unpack into four directories, test containing the images in Test Set A, test-low contains Test Set B, newtest contains Test Set C, and rotated contains the Rotated Test Set.
Henry A. Rowley (har@cs.cmu.edu)
Shumeet Baluja (baluja@cs.cmu.edu)
Takeo Kanade (tk@cs.cmu.edu)