A device may detect, in a field of view of a camera, one or more components of an automated teller machine (ATM) device using a computer vision technique based on generating a three dimensional model of the one or more components. The device may identify the ATM device as a particular device or as a particular type of device based on the one or more components of the ATM device, or first information related to the ATM device. The device may identify a set of tasks to be performed with respect to the ATM device. The device may provide, for display via a display associated with the device, second information associated with the set of tasks as an augmented reality overlay. The device may perform an action related to the set of tasks, the ATM device, or the augmented reality overlay.