Dataset of Public Objects in Uncontrolled Environment for Navigation Aiding

Computer vision is a new approach to navigation aiding that assists visually impaired people travel independently. A deep learning-based solution implemented on portable device uses monocular camera capture public objects could be low-cost and handy aid. By recognizing in the street estimating their distance from user, are able avoid obstacles outdoor environment walk safely. In this paper, we created dataset of an uncontrolled for aiding. The contains three classes which commonly exist pavements city. It was verified high quality object detection estimation, ultimately utilized as aid solution.