The objective of this master thesis is to improve the quality
of face detection and recognition by combining existing
software solutions. In particular, four pieces of face software
have been chosen and tested, as well as several ways of their
combination. The specific combination proposed in this thesis
has significantly improved the quality of face detection.
Another combination has been proposed for face recognition with
the aim at utilizing advantages of individual software
solutions. Such combination has also achieved the best results.
The proposed approaches have been implemented as two Java
projects. These projects provide users with a possibility to
combine an arbitrary number of software for face detection and
recognition.