Privacy-compliant Metadata Mining
NATIX Network uses proprietary computer vision AI to analyze video streams, turning visual data into privacy-compliant metadata. NATIX’s unique privacy-by-design technology relies on two main pillars:
- Edge Computing Architecture (instead of centralized Cloud) - By enabling local computing (i.e., on-camera or on-device), the need for transferring raw images to data centers is eliminated. The AI will analyze the video locally to mine “metadata,” i.e., numeric data such as the number of detected cars and their location. The information shared with the network is the mined metadata. This architecture has both privacy protection and bandwidth benefits for the network.
- Anonymized Detection (free of personal data) - Faces, skin color, vehicle number plates, and other personal information are irreversibly discarded. Computer vision models will perform their detection over anonymized images free of personal data.
This framework turns any camera anywhere into a smart privacy-compliant data source that detects and transfers numeric-only information.
Metadata mining - Image and video data is analyzed by computer vision AI to create numeric metadata (e.g. number of detected cars and their location). This process is referred to as metadata mining.
Last modified 1yr ago