Real-time biometric face recognition and pedestrian / vehicle tracking SDK for viodeo surveillance systems and networks. Automatic identification against watch list. Vehicle make and model estimation. Pedestrian mask, cloth and gender estimation.
Version: 8.1Real-time biometric facial recognition and pedestrian or vehicle tracking in live video streams.
License: Free To Try $865.00
Operating System: Windows
Available as SDK (software developing kit) for technology integration into small and large surveillance systems.
The proprietary technology provides automatic operation with multiple face detection, tracking, persons' gender determination, mask wearing detection, enrollment and identification.
Vehicle make and model estimation. Vehicle orientation angle estimation. Automatic license plate (number plate) recognition (ALPR / ANPR).
Cloth and gender estimations for pedestrians, including face mask and partial or full nudidity detection.
Can be used in law enforcement, security, attendance control, visitor counting, traffic monitoring solutions.
Some applications may require to enroll the users in advance (such as list of employees or watch-list of unwanted persons or suspect criminals), whereas other just work on the general public passing by (i.e. detect if passers-by draw attention to certain ads or billboards).
The SDK allows to create networked applications and systems with multiple cameras and data processing nodes on Windows and Linux platforms.
Includes programming samples in Java, .NET, C, as well as all necessary components (the biometric technology, camera support, network synchronization).
The trial requires constant Internet connection.
Version 8.1: Updated face detection algorithm, working better with cropped faces, masked faces, bad pose faces. Improved face blink, dark glasses, expression, gender, glasses, mouth open, beard, mustache, hat attribute detection. Introduced heavy frame detection. Fixed NSurveillanceEngine processing error.
Version 8.0.2021.11.25: Algorithms-related changes: Clothing details now also determines clothes by body parts: headwear, torso, arms, legs, feet. Improved vehicle model prediction. New features: New Java sample: surveillance-sample. Fixes: Fixed a number of minor bugs, crashes and issues.
Version 8.0.20210512: API for easy selection of the best frame added. API for retrieving subject information for all tracked frames added. New Java sample: surveillance-images-sample. New C++ sample: SurveillanceImagesSampleWX. Fixed a number of bugs and issues. Multiple minor improvements.
Version 8.0.2021-05-12: New features: surveillance-images-sample - a new Java sample. SurveillanceImagesSampleWX - a new C++ sample. Added an API for easy selection of the best frame. Added an API for retrieving subject information for all tracked frames. Other improvements and bug fixes.
Version 8.0: New face detection and recognition of people who are wearing masks. New algorithms that improve license plate detection and recognition speed and accuracy. Vehicle make and model estimation. Vehicle orientation angle estimation. Cloth and gender estimations for pedestrians.
Version 7.3.2020.11.30: Fixed face medium template compatibility with MegaMatcher Accelerator. Improved SurveillaceFacesThermalSampleCS. Reduced GPU memory consumption for faces modality.
Version 7.3.2020.09.16: Fixed issues with event time stamps when reading from video file. Fixed a number of licensing-related bugs and issues. Fixed issues with diagnostic report.
Version 7.3.2020.03.02: Improved License plate recognition (ALPR) algorithms. Fixed a number of bugs and issues. Lots of improvements to SDK and samples.
Version 7.3: Improved Vehicle-Human and License plate recognition algorithms Added functionality to use still images with Vehicle-Human and License plate recognition modes Added SurveillanceImagesSampleCS and DetectFromImage tutorials Fixed licensing related issues Multiple GPUs (up to 8) now supported