Crime prevention, force protection, and intelligence operations all rely heavily on video and imagery data. However, the number of files of this nature greatly exceeds that of available analysts, making manual review expensive and inefficient. What’s more, with the rate at which video and imagery data is captured and available for analysis, an additional challenge becomes identifying actionable intelligence before it’s too late.
Enhance and expedite video and imagery exploitation efforts with Advanced Video Activity Analytics (AVAA), an automated analysis solution designed to increase the rate of actionable intelligence produced from massive volumes of unprocessed media. With a scalable and standards-based framework, multi-source intelligence access, and real-time alerts of emerging threats, AVAA is the solution that not only greatly reduces analyst workloads but also saves valuable time and funding.
For more information, download our datasheet.
AVAA provides state-of-the-art video exploitation when compared to other available options. It fosters many video and metadata processing capabilities, from real-time and forensic viewing to indexing and scalable storage. Ingested FMV streaming data can be searched and retrieved geospatially, temporally, or by annotation and then exploited via advanced video analytics ranging from georegistration and stabilization to face and license plate detection. Once actionable data is extracted, analysts can create video analysis products such as clips, snapshots, and annotations for dissemination to other analysts via real-time, web-based collaboration.
AVAA offers a standards-based framework to integrate new video analytics, combining commercial, government, academic, and open-source computer vision algorithms by making them interoperable and scalable. Additionally, it provides a Software Development Kit (SDK) to facilitate the integration of emerging algorithms and their interfaces, ensuring that AVAA remains a standardized solution.
Through the Video Analytics Web Services (VAWS), AVAA enables existing systems to quickly take advantage of its powerful analytics. All entities extracted from video—from faces to vehicles—are captured in AVAA’s Video Data Model (VDM) and can be discovered and viewed through AVAA’s user interface or made available through queries against VAWS.
AVAA integrates with several components, increasing its value as an all-inclusive, rapid exploitation solution.
- Video Processing and Exploitation Framework (VPEF), a modular software architecture providing a computer vision algorithms standard, an SDK for integrating new algorithms, and a GUI for addressing pipelines comprised of individual computer vision plug-ins
- Video Data Model (VDM), a model that captures information from multiple levels of FMV processing, the evidence chain from the original pixels, and manual analyst annotations
- VDM Annotation Web Service (VAWS), a service offering methods for querying and returning data from the AVAA video data store to external systems as well as software for manual annotations, automatic georegistration, and data storage
- Cloud Analytics Collaboration Environment (CACE) UI Framework, a web-based GOTS framework that integrates capabilities from disparate sources into a single common UI for real-time collaboration, alerting, data sharing, and API searching
- AVAA VM-based System, an end-to-end test bed for capability integration