AXIS Scene Metadata
Intelligent Video Surveillance at the edge
AXIS Scene Metadata is an advanced AI-powered analytics solution that enhances video surveillance capabilities by providing detailed insights about objects and events in a scene. Designed for use with select Axis camera models equipped with machine learning or deep learning processing units, this technology significantly improves video searchability, automates security operations, and enables more efficient decision-making. By generating rich metadata directly at the edge, AXIS Scene Metadata offers enhanced scene understanding, rapid forensic searches, and seamless integration with video management systems, making it a valuable tool for modern security and surveillance applications.
Core Functionalities
AI-powered scene analysis is at the heart of this technology, which detects and classifies objects such as humans, vehicles, and faces in real time. The system generates detailed metadata about these objects, including their position, speed, direction, and specific attributes like vehicle colour or clothing type. This edge-based processing occurs directly on compatible Axis cameras, reducing server load and improving system efficiency. The generated metadata is then seamlessly integrated with video management systems and other security solutions through open APIs and ONVIF Profile M compliance, enabling advanced search capabilities and automated actions based on predefined rules.
Improved Video Searchability
This technology uses AI-powered analytics to transform raw video data into highly searchable content. Users can perform real-time and forensic searches based on specific object attributes like colour, type, or location, significantly reducing the time needed to locate relevant footage. The system’s integration with video management systems (VMS) enables efficient filtering and retrieval of video segments using precise search parameters. This enhanced searchability streamlines investigations accelerate decision-making processes and improves overall operational efficiency in surveillance systems.
Detectable Attributes
Object detection capabilities include humans, faces, vehicles (cars, buses, trucks, bikes), and license plates. For cars, the system can identify colour and license plate information. Human attributes include upper and lower clothing colours. General attributes detected across objects are position, speed, direction, confidence level, distance, and geo-coordinates (longitude and latitude). This granular level of detail enables precise scene analysis and supports a wide range of security and operational applications.
Security Operation Benefits
Enhancing security operations, this technology offers numerous advantages, including automated actions triggered by predefined rules, improved situational awareness, and rapid decision-making capabilities. Processing data at the edge reduces central server dependency and minimizes data transmission costs, leading to significant operational cost savings. The system’s open API and compliance with standards like ONVIF® Profile M enable easy customization and scalability, allowing users to tailor solutions to specific security needs and expand as required. These features collectively contribute to a more efficient, responsive, cost-effective security ecosystem.
Call us on 1300 556 334 or email [email protected] to learn more.
Customers in New Zealand call 0800 345 677 or email [email protected].