Welcome to the Face Detection Database of “Universidad de las Fuerzas Armadas – ESPE”

This page contains a set of images to train and test a Face Detection System in unstructured environments. This data set includes 7117 color images in jpg format

This Database is composed of the following three folders:

Training Database

This folder contains 5750 images of 64×64 pixels each for training. In this folder, 2750 images contain faces with a frontal pose and the remaining are background images.

Testing Database

Testing2 Database

These two folders contain 1367 testing images. The test of a detection system is divided in two stages: classification and detection. For testing the classification there are 959 color images of 64×64 pixels each. For testing the detection system there are 408 color images of 640×480 pixels each.

If you use this database, please cite the following paper:

Jonathan S. Talahua, Pablo S. Aulestia, Victor H. Andaluz, and Marco E. Benalcázar, “Real-Time Object Detection using Artificial Neural Networks”, Submitted for review, 2017