Design of Mobile Augmented Reality Assistant application via Deep Learning and LIDAR for Visually Impaired

In daily life, finding lost items is one of the biggest challenges for the visually impaired. Due to their vision problems, finding the target item is difficult, especially when they are not in the same space. Many navigation devices have been designed to assist the visually impaired in real life. But these devices require additional purchases, lack flexibility, and cause inconvenience to visually impaired users. In this research, we designed an application named NAAD on a mobile device with a LIDAR function to achieve obstacle avoidance, item query, and indoor navigation via Deep Learning and AR technology to improve the quality of life for the visually impaired. This paper will focus on a new approach that integrates object recognition, route planning, and target object navigation. With this new method, the NAAD application can guide users to find user-specified lost items in the same or different spaces.