Yasmina Andreu Cabedo


Online Study of Gender Classification from Facial Images

This survey investigates the accuracy with which humans can recognise gender from face images. For each image you see, you will be asked to choose if it is a male or a female. Normally, the images will contain small-sized faces, although sometimes other types of images can appear due to errors in the face detection. If you come across an uncertain image, you should choose the gender that seems most appropriate. Some images show people wearing sunglasses or a scarf to make the gender recognition harder. In all cases, just provide what you believe to be the best answer.

The number of images in this survey is quite high, you should keep in mind that you don't have to label all the images in one session. The website allows you to save your answers and continue later.

The results obtained in this survey will be compared to those achieved by automated systems, so we can check who solves the problem for recognising gender better, humans or machines. You don't have to worry about making mistakes, although you should always try to give the answer that you think is correct. Making errors intentionally could affect the results of the study we are carrying out.

Participate in the study

For any questions contact: yandreu(at)uji.es