BAUTA

01/06/2021 to 31/05/2022

We aim to make optical data collection not only data protection compliant but completely anonymous! With our developed blind sensors and the use of Deep Neural Networks we want to enable adaptive marketing in the analog "Digital out of Home" advertising sector.

A prerequisite for adaptive marketing in the physical shopping world is to capture optical data. Our sensors have already been confirmed in this regard by a legal assessment in their complete data protection, with specially trained DNN object recognizers for their anonymous image and submitted to the DPMA for a patent. The further procedure involves intensive collaboration with the University of Tübingen in two areas: First, together with the Chair of Cognitive Systems, the amount of the amount of recognizable information beyond the freely accessible state of AI technologies.
will be expanded. Thus, in addition to common recognizers (age, gender, position), we also want to train innovative object recognizers, such as an optical detection of emotions via body posture or the recognition of attention and or the recognition of attention and motion vectors. Second, in the collaboration with the Chair of Marketing the use of practical data and the sum of the recognized information will be used to define customer types by means of cluster algorithms, which have show similar interests. We plan to improve existing research findings through our gained information. In parallel, with the help of our external mentors and supporters, we will use our grant funds to finalize our sensor
board layout, in order to be able to offer a complete IT infrastructure for adaptive marketing at the end of the EXIST period.