Contribution to the fusion of soft facial and body biometrics for remote people identification

In this work, we are interested in exploring soft biometrics to identify people at a distance without recourse to their cooperation. The proposed system is based on fusing two soft modalities of the face, which are the skin and hair colors, and six soft modalities of the body — one for the height, four for the body measurements, and one for gait cycle. Tested on a Chimera database of faces and video sequences, the fusion at the decision level using the technique of majority voting of soft facial and body traits has led to a 95% recognition rate.