Magic Vision provides Silver Eye, an AI-powered computer vision system that automatically detects falls and monitors the well-being of elderly or vulnerable individuals in residential care settings. The platform operates entirely on-edge, processing video locally to ensure instant response times and complete privacy protection without sending sensitive footage to the cloud. It also offers customizable access management, behavioral anomaly detection, and configurable notifications for caregivers via app, email, or dashboard.
Funding
Funding not disclosed
Founders
Product
Problem
Elderly individuals living alone or in care facilities face significant risks from falls, which often go undetected for extended periods, delaying critical intervention. Traditional monitoring solutions rely on wearable devices that users may forget to wear or reject, while camera-based systems often raise privacy concerns and generate numerous false alarms that desensitize caregivers.
Solution
Silver Eye by Magic Vision is an AI-powered computer vision platform that automatically detects falls and monitors behavioral patterns without requiring any wearable device. The system analyzes movements in real time to distinguish genuine falls from normal activities, significantly reducing false alarms and enabling rapid response. All video processing occurs locally on the device using edge AI technology, ensuring instant reactions, complete data privacy, and reliable operation even during internet connectivity issues. Caregivers receive customizable notifications through app, email, or dashboard, with alerts that redirect directly to event pages for quick severity assessment. The platform also monitors bathroom usage frequency, time spent in bed, apartment presence, and prolonged inactivity, providing a comprehensive well-being overview while respecting privacy through configurable video streaming options.
Target Audience
Primary customers are assisted living facilities, home care providers, and families caring for elderly individuals who need continuous, non-invasive monitoring to ensure safety and timely intervention in case of falls or unusual behavior.
Features
- Real-time fall detection using computer vision algorithms that differentiate critical events from false alarms, including collapses and falls from beds or sofas
- On-edge AI processing that analyzes video locally on the device, eliminating the need to send sensitive footage to the cloud
- Privacy-preserving video configuration with options for clear, blurred, skeletal, or event-log-only storage modes
- Access management system that monitors entries and exits from safe areas, detects unknown visitors, and verifies resident presence in the apartment
- Behavioral monitoring that tracks bathroom usage frequency, bed occupancy time, apartment presence, and prolonged inactivity periods
- Customizable notification system delivering alerts via app, email, or dashboard, with direct links to event pages for severity evaluation
- Night vision capability ensuring continuous monitoring even in dark environments
- No wearable devices required, making the system suitable for users who may forget or refuse to wear tracking devices