Soft biometrics such as gender, height and weight, unlike hard biometrics, do not have the capacity to differentiate between two different identities but can be useful to improve the quality of different biometric systems.

Among those, weight is also an indicator of both physical aspect and health conditions and in opposition to gender and height, body weight changes during the adult life of a person and needs to be periodically measured.

Conventional weight measurement techniques require the cooperation of the subject to be measured, which might not be possible during medical emergencies, road accidents or due to different patient disabilities.

This database is composed of 400 (304 male and 98 female) frontal face images.

The database includes annotation of the gender, age, height, weight, ethnicity, gender, eyes and hair color of the subjects present in the database.

The images and annotation presented in this database were collected from the Polk County Jail official webpage. All parameters were computed at the same moment as the face picture was taken, therefore guarantying the accuracy of their annotation.

Any publication reproducing partially or totally this database must cite the following paper:

Mirabet-Herranz, N., Mallat, K., & Dugelay, J. L. (2023, January). New Insights on Weight Estimation from Face Images. In 2023 IEEE 17th International Conference on Automatic Face and Gesture Recognition (FG) (pp. 1-6). IEEE.

@inproceedings{mirabet2023new,
title={New Insights on Weight Estimation from Face Images},
author={Mirabet-Herranz, Nelida and Mallat, Khawla and Dugelay, Jean-Luc},
booktitle={2023 IEEE 17th International Conference on Automatic Face and Gesture Recognition (FG)},
year={2023} }