Face Analytics
SAIMOS® Face Analytics effectively and reliably detects and recognises faces. The simple rule system offers many possibilities to implement face analytics throughout various industries. For example, Face Analytics can be used for access control, backlisting of shoplifters, etc.
SAIMOS® Face Analytics uses state-of-the-art AI systems to detect, recognize and compare faces. It’s optimised for Intel® CPU/GPU. The user-friendly interface provides step-by-step configuration enabling fast setup times as well as alarming and reporting.
Potential use cases specialise on real time alerting (based on black-/whitelist) and flexible reporting of detected and recognized faces as. Also, forensic analysis for investigations are supported.
Now you can fully use SAIMOS® Face Analytics from within Milestone XProtect® by simply installing our
SAIMOS® Video Analytics Plugin for Milestone! The system is fully integrated into Milestone XProtect® Essential+ or higher and can be configured directly from within the Milestone XProtect® Management Client. The simple scene calibration and rule definition allows an initial configuration within just a few minutes.
Face features are stored as binary vectors from which faces cannot be reconstructed and thus supports personal data protection.
Use Cases
| Access Control | Face Analytics in its one to one comparison mode complements access control systems. Our REST API allows a flexible integration. Custom integrations for access control systems are possible as well. |
| Black-/Whitelisting | Face Analytics supports Black-/Whitelisting to e.g. secure buildings, stadiums, shops or homes. Faces added to black- or whitelists will trigger an event. Based on such events, an action can be triggered when e.g. a VIP or an unwanted person is detected. |
| Forensics | Face Analytics in its forensics mode enables the search for suspects in archive material and optimizes the investigation time. Face Analytics Pro can use a live face database, VMS system or exported videos as input. The right setup will reduce investigation time from days to seconds. |
| Retail | In retail, it’s worth to know your customer’s retention times but also the number of re-visits over a certain period of time. Further, it’s useful to know the customer segmentation by gender and age over time. In addition, potential shoplifters can be blacklisted to inform the security when such a person enters the premises. |
| Investigation | Criminal investigations can be enhanced by using a combination of live and forensic mode. The live mode will detect faces and use the blacklist mode to trigger alerts while the forensics mode will use the collected live face database to create detailed reports. |
Key Features
| Algorithms |
|
| Configuration |
|
| Stream Integration |
|
| Events |
|
| Technical Requirements |
|