EyeKnow

EyeKnow has developed a first-of-its-kind artificial intelligence technology for detecting violent behavior by caregivers toward toddlers and elderly people, based on video and advanced algorithms for analyzing movements and interactions. The platform connects to existing security cameras (CCTV, DVR, and IP cameras), collects video in real time, and runs deep learning models trained on original violence footage to identify aggressive actions such as forceful pulling, pushing, force-feeding, hitting, and threatening hand movements. The system sends alerts about violent incidents to a dedicated app, and is supported by a human-in-the-loop monitoring team that verifies alert accuracy, resulting in especially high accuracy rates and an almost zero false-positive rate.

The architecture is built on three main components: camera integration and data collection, data processing and algorithm execution, and alert delivery to customers when unusual incidents are detected. On top of the detection layer, EyeKnow is also developing a BI and dashboard layer for broad monitoring, tracking behavioral patterns, and improving the professional performance of caregiving teams. The company is planning two patents: one for the ability to detect aggressive movements toward vulnerable individuals, and another for identifying caregiver profiles with a high likelihood of committing violence, enabling early detection and preventive monitoring. The technology is aimed at vast private and institutional markets – from nursing homes, assisted living facilities, and in-home elder care to kindergartens, daycare centers, and special education – with potential expansion into violence prevention in public spaces.