Percipient.ai logo


Percipient.ai Solutions for the Public Sector

Percipient.ai is a Silicon Valley-based company focused on creating and delivering visual intelligence products to address critical national security and intelligence missions. Mirage™ is our leading multi-source, intelligence analysis platform that triages large amounts of full motion video, still imagery and geospatial data for ground, air and space missions. Mirage provides simultaneous person, vehicle, and object of interest live recognition, historical analysis, and internal authorized person and specific vehicle monitoring for a defense-in-depth and comprehensive force protection posture. The fixed and mobile appliance-based platform uses an intuitive workflow to discover patterns of life, identify networks, and provides alerts/recognition for users on people, vehicles, and custom objects – with precise zoning, all essential capabilities for providing customers with relevant and actionable alerts of potential threats.

  • Full Motion Video Module (FMV)

    FMV couples collaborative and intuitive workflows with facial recognition and correlation, vehicle recognition, persistent zoning, and object detection for pattern of life analysis and geo-correlation of results in video, still images, and live streams. Out of the box come a dozen plus detectors. Should new custom object requirements surface, users can independently create them using our separate MINT custom object capability.

  • Geospatial Module (GSM)

    GSM allows for collaborative teaming and analysis across GEOINT missions. GSM works with the user to confirm detection, identification, classification and characterization of objects and vehicles across geospatial data sources. Users can also geo-fence a zone and set persistent overwatch and alerting. Using GSM, GEOINT users confirm historic data and link satellite imagery feeds into Mirage for processing, alerting and analysis.

  • Vehicle Recognition Module (VRM)

    VRM enables users to establish situational awareness while supporting target detection, tracking, recognition, identification, and determination of hostile intent. VRM is a native application capable of analyzing moving vehicles for detection of surveillance behaviors and matching observations, including license plate data and other embeddings, to detect and identify vehicle patterns and reveal surveillance.

Core Capabilities:

Mirage provides rapid triage and analysis of both live and historic full motion video and still imagery for person, vehicle, and object characterization. Linking and correlating a wide range of signals or geolocated data – whether IMSI/mobile, .kml/.kmz data layers, or other data from ELINT sensors or map collected reference emitters, Mirage will map, correlate, and present relevant data allowing the analyst the ability to uncover latent patterns and relationships hiding in massive data sets. Users save time, reduce misses and errors, and open up previously untapped or under-resourced data sources. Employing automated or manual reporting tools, intelligence operators can share their findings amongst teams, broader units, or higher echelons. Models are strengthened by users’ individual confirmations and conclusions are shared across the whole organization, improving work products, and generating enterprise intelligence.

  • Full Motion Video (FMV) Key Features:
    • Person Recognition: Analyze videos, still images or monitor live feeds for specific identities that have been uploaded
    • Alert on Person(s) of Interest: Alert when a known or unknown person appears in a live stream
    • Vehicle Recognition: Analyze videos for specific vehicle identities
    • Object Detection: Search for backpacks, handbags, cars, buses, motorcycles and more, including custom object models created in the MINT module deployed with FMV
    • Zone & Loitering Monitoring: Create user defined persistent zones and loitering time for alerting in historical videos or live feeds
    • Identity Creation/Package: Discover and create persons of interest. Associate identities and proxies that are in the same network together and query for their presence across data
    • Network Discovery: Discover potential target associates through on-the-fly identity creation and querying for a correlation to a known person of interest
    • Collaboration: Create, share and export queries, reports, and identity packages for greater collaboration across teams
    • Simultaneous Real Time Monitoring: Analyze up to 10 active live streams per instance that can be private per user or shared among teams
    • Automatic learning: Confirm hits on persons of interest to an identity package which automatically strengthens the accuracy of the search
    • Model Integration: Incorporate 3rd party deep learning models such as TensorFlow or MxNet to recognize objects not already present in Mirage out of the box
    • Confidence Thresholding: Adjust the confidence threshold level so the system alerts on only high confidence recognition or includes lower confidence recognitions when querying with a weak or dated identity package
  • GeoSpatial Module (GSM) Key Features:
    • Analyst workflow: Supporting detection, classification and geo-correlation of objects from airborne and satellite imagery
    • Optimized human + machine teaming: Designed to build trust
    • Confirm/Reclassify/Reject: To generate training data and analyze imagery
    • Automated learning and strengthening of classification models: Delivers constantly improving mission accuracy
    • Intuitive geo-fencing and automated alerting: For tipping and cueing and object detection
    • ELT features: Haze removal, contrast adjustments, measure tool, etc.
    • Open architecture: System that operates at speed and scale
    • Organic labeling and integration with other CV models
    • Pattern of life detection
    • Geo-correlation
    • Neural network models
    • Export and import of models: Models and training data exportable from low to high side
  • Vehicle Recognition Module (VRM) Key Features:
    • Live alerting: To potential vehicles of interest (VOIs)
    • Geo-correlated characterization: Track VOIs across missions
    • Comprehensive understanding in real time
    • Integrated license plate recognition
    • Identify and confirm vehicles of interest
    • Real-time vehicle and route tracking
    • Integrated person and objects of interest analysis

Technical Specifications

With its open architecture, Mirage is highly extensible, allowing for ease of integration into other systems and modifications based on evolving mission requirements. The components of the platform are decoupled microservices, using Apache Thrift and REST APIs for communicating between them. These components are also Dockerized, allowing use with container orchestration platforms, such as Kubernetes, to scale out and parallelize the processing, deployable on the cloud or on-premise. Using state-of-the-art deep learning algorithms, the platform can run on both CPUs and GPUs. Mirage’s core framework can easily be extended for other forms of structured and unstructured intelligence data.

Demo
Contact our mission team (info@percipient.ai) to schedule a demonstration of our software.