Generating actionable metrics from sensor data

Abstract

This disclosure is directed to systems and techniques for generating actionable metrics data based on sensor data generated from one or more sensors within an environment. For instance, a brick-and-mortar retail environment or other materials handling facility may include one or more sensors, such as overhead cameras, which generate image data as users interact with items in the environment. Machine-learning models and/or computer-vision algorithms may then analyze the resulting image data to identify the interactions between the users and the items. The system may then generate metrics data indicating varying metrics associated with these interactions, which may be used to generate graphical user interfaces (GUI) for presenting the metrics data to vendors, brand owners, and the like.