GAP-AI, by Synesthesia, can be labelled as an optimization system for retail processes programs. Nowadays, this activity is time-consuming and likely to generate errors, since it requires workers to visually check the shop shelves, evaluating: exhibitive situation, products availability, positioning, promotions, contractual agreements with suppliers. The application of Artificial Intelligence is strongly pushing forward the automation of this process. The picture of the shelves, taken with smartphones and tablets, are sent to the Cloud, where they are elaborated so that the products availability and positioning are calculated. One of the major limitation of the available solutions is the processing speed.

The recent product available in the market need several minutes before the validation of correct acquisition of the pictures, together with the generation of some results. Synesthesia is trying to investigate some speed improvement, distributing the AI analysis between the smart device and the cloud, reducing the process time, and optimizing the operations for the assigned worker. More in details the main objective of the project is the definition, realization and testing of a Proof of Concept of the proposed solution for the pictures’ analysis, based on convolutional neural networks, implemented and optimized between terminal and cloud.