Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
Computer Vision for Public Transportation - Phase 1
Overview Solution overview Partner overview
Partner overview
The EPIC iO Universal IOT Gateway
Real-time algorithms and insights
EPIC iO Cloud
More information
Lab architecture overview Use case validation Conclusions and References

Real-time algorithms and insights

Home > Workload Solutions > Computer Vision > White Papers > Computer Vision for Public Transportation - Phase 1 > Real-time algorithms and insights

Real-time algorithms and insights

  • DeepInsights is an open AI + IoT software suite that combines and enriches video, external feeds and sensor data into a seamless operational experience including:

    • Trends and insights
    • Alerts to enterprise applications
    • Live and recorded video monitoring
    • Deployment and management of cameras, sensors, and edge devices at scale
    • AIoT open API
    • Hybrid cloud design including edge, data center, and/or cloud architecture

    DeepInsights integrates AI models, automated logic, and visualization to deliver actionable insights faster with less effort. DeepInsights AI models can be configured to detect and track objects as well as define human posture. Posture definition can also be configured to generate alerts if a person is detected moving quickly from a standing to a sitting or laying position to evaluate distress or nefarious behavior.

    Video processing with DeepInsights starts with the ingestion of real-time streaming video (RTSP) streams that then get analyzed using object detection, object tracking, and other enhanced analytics for each frame. DeepInsights can process live video or legacy recorded video retrospectively and has complete feature parity with GPU (Intel CPU + NVIDIA or Intel CPU + Intel Flex GPU) as well as CPU-Only (Intel CPU only).