ALO: AI for Least Observed People

• Autor(a) principal: Mamun, Shamim Al
• Data de Publicação: 2021
• Outros Autores: Daud, Mohammad Eusuf, Mahmud, Mufti, Kaiser, M. Shamim, Rossi, Andre Luis Debiaso [UNESP]
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1007/978-3-030-82269-9_24; http://hdl.handle.net/11449/222288
• topic: Blind vision; Face recognition; Object detection; Smart glass

In recent years, visual assistants of humans are taking place in the consumer market–the eye-line of humans equipped with a see-through optical display. Computer Vision Technology may play a vital role in visually challenged people to carry out their daily activities without much dependency on others. In this paper, we introduce ALO (AI for Least Observed) as an assistive glass for blind people. It can listen as a companion, read from the internet on the fly, detect surrounding objects and obstacles for freedom of movement, and recognize the faces he is communicating with. This glass can be a virtual companion of the users for social safety from unknown people, reduce the dependency of others. This system uses the camera for identifying human faces using MTCNN deep learning technique, bone conduction microphone, and google API (Application Programming Interface) for translating voice to text and text to bone conduction sound. A Market Valuable Product (MVP) has already been developed depending on our survey of over 300 visually impaired persons in Europe and Asia.