The approach for automatically adjusting display setting for a user that uses corrective lens or not is disclosed. The approach determines whether a user is wearing corrective visual lens. If the user is not wearing correct visual lens, then the system measures the ocular features of the user via sensors. The system then transmits data associated with the ocular features to a deep learning server wherein the deep learning server analyzes the data. The server generates a customized display setting based on the analyzed data. The system receives the display setting from the deep learning server and automatically adjusts the display setting on a device of the user based on the received display setting.