We would like to inform about release of VeriLook Surveillance 1.1 SDK for developing software that performs passive biometric face identification using live video streams from digital surveillance cameras.
The latest version includes a new facial recognition algorithm and features these updates:
* Increased recognition accuracy, i.e. decreased False Rejection Rate, by 10-50% at the same False Acceptance Rate level with respect to the VeriLook Surveillance 1.0 algorithm.
* Recognition of faces which are rotated in-plane (rolled/tilted left or right) at any angle (the previous version could detect only faces with ¡¾15 degrees roll angle deviation).
* Improved facial template generalization, enabling a more reliable template to be generated from a set of previously created templates.
* VeriLook Surveillance 1.1 templates are fully compatible with VeriLook 4.0 templates.
* Updated samples for all programming languages.
* Added C++ sample.
* Added Mobotix DualNight M12 IP camera support module.
* Video files support module added.
* C++ wrapper for SDK functions added.
Algorithm demo applications and 30-day trial version of VeriLook Surveillance 1.1 SDK are available for download.
Existing VeriLook Surveillance SDK customers receive a free upgrade to the new version of the SDK.
For more information, please visit this webpage at http://brucenbrian.com/product1_7_11.html