A SURVEY ON RECENT ADVANCES IN AI AND VISION-BASED METHODS FOR HELPING AND GUIDING VISUALLY IMPAIRED PEOPLE

Abstract : We present in this paper a state of the art and an analysis of the recent research work 1 and achievements performed in the domain of AI-based and vision-based systems for helping 2 blind and visually impaired people (BVIP). We start by highlighting the recent and tremendous 3 importance that AI has acquired with the use of convolutional neural networks (CNN) and with 4 their ability to efficiently solve image classification tasks. We remind also that VIP have great 5 expectations about AI-based system as a possible way to ease the perception of their environment 6 and to improve their everyday life. Then we set the scope of our survey: we concentrate our 7 investigations on the use of CNN or related methods in a vision-based system for helping VIP. 8 We analyze the existing surveys and we study the current work (a selection of 30 case studies) 9 along several dimensions such as acquired data, the learned models and the Human-Computer 10 interfaces. We compare the different approaches and we conclude by analyzing the future trends 11 in this domain.