Object Detection with Audio Comments using YOLO v3

Vision loss is a serious factor that can affect one's course of life and quality of life, making them barred from independence, easy mobility, unnecessary tripping over objects, etc. As per World Health Organization's most recent studies, there are around 2.2 billion people on the planet who are visually impaired, and a major proportion of the population is over the age of 50 years. The primary cornerstone of this paper is about how a simple Object Detection model, done with the help of some python libraries and APIs, shall prove to be helpful for visually impaired people, and help with those subtle challenges. In this proposed methodology, to overcome the aforementioned drawbacks, a Machine Learning model whose key purpose is to detect the location of the object surrounding the person and provide voice feedback using Google Text-to-S peech API has been implemented. To train the model around various aspects of the object, the MS -COCO dataset is being used.