SELF LEARNING FACE RECOGNITION USING DEPTH BASED TRACKING FOR DATABASE GENERATION AND UPDATE

摘要

This over time and individuals that are providing examples of commonly performed facial recognition training database generation techniques to collect, including personal characteristics of the face resulting from the movement in the environment to create a training database of facial characteristics for individual fraud. Since the face attribute captured in time, it will exhibit the observed under a variety of angles and distances, different resolutions and different environmental conditions (e.g., lighting conditions, and turbidity) individual face. In addition, for a long period of time the characteristics of the individual's face can be collected periodically, these characteristics may exhibit a change in the appearance of the individual. This produces a rich training resource for use in face recognition systems. In addition, because of the personal face recognition training database may need to be established before by the face recognition system, further training will speed up when in use. ;