iSight is an application designed to help the visually impaired community better understand their surroundings and also assist in doing their day-to-day tasks easily.
The app has voice-based features like:
- Scene detection: Describe your surroundings like identifying furniture, scene/ room type.
- Face detection: Identify when your contacts (pre-defined) are around and their emotions.
- Share a first-person view with a contact so that they can see what you are seeing and guide you manually.
- Product identification and detailing: Scan a barcode to identify a product using Google search and announce product name/ colour to assist in buying.
- OCR: Read text like road signs, menu, or product description.
It’s built for the Vuzix Smart Glasses and makes use of the following integrations:
- Azure Cognitive Services
- WebRTC using PeerJS for Video calling
- Alexa Voice Service
- Google Search API
The app opens up with a camera preview to allow capturing the surroundings. The users can select various features using voice commands and start the snapper to start capturing the surroundings. All the responses are converted into speech and spoken to the user.