TensorSight - Vision Assistant

by A.Q


Health & Fitness

free



Do you suffer from visual disabilities or know someone who does?Do you also know that with the advances made in Artificial Intelligence it is possible for robots to literally “see”?The good news is that we can use the same technology to alleviate the problems faced by the visually impaired community.TensorSight was developed to serve as a free visual aid for the visually impaired community.How you may ask? It’s simple. TensorSight is an Android app controlled primarily through voice commands and gestures. Smartphone vibrations are used to indicate the presence of widgets on the screen. The app features are listed below.FEATURES:Obstacle Detection: User can detect obstacles in the visible environment simply by pointing the smartphone camera in any direction. Only recognized obstacles are narrated to the user. Currently, TensorSight supports only a hundred common obstacles and also narrates the colour of an obstacle. The nearest obstacle is also specified.Text Detection: TensorSight narrates any Latin-based text to the user. This feature is intended for short texts; however, the user can also read long documents if they fit in the camera view. A common practice would be to centre the smartphone camera parallel to the centre of the document at approximately an arm-length distance.Person Detection: TensorSight detects people in the camera view and specifies whether an individual is smiling or is asleep. (The functionality of these features will be extended in the future)Barcode Scanning: TensorSight detects barcodes in the camera view and indicates their detection to the user. The information encoded by the barcode is also narrated to the user, and for barcodes with a stock-keeping unit (SKU) product verification is also done.Still not convinced at the potential of TensorSight to help you or your loved ones? Try out the app. It’s free! And don’t forget, we give the vision!--------------------------------------------------------- We would love to hear from you! If you have feedback, questions, or concerns, please email us at: [email protected] or follow us on Facebook: https://www.facebook.com/tensorsight ---------------------------------------------------------