Real-Time AI Processing Framework for Video and Multi-Sensor Applications



kvSonata :

  • Easy-to-use real-time video and multi-sensor processing framework

  • Develop in Python, C++ or Golang

  • Quickly prototype, develop, test and deploy

  • Container-based, scalable architecture

  • Many use-case examples provided to get you started

  • Supports multiple machine learning frameworks

  • Supports standard and optimized deep neural network models

  • Real-time stream processing with web-based visualization

  • Compatible with GPU, VPU and CPU hardware

  • Easily create custom processing modules

  • Time-synchronization of multiple input data streams

  • Run multiple independent user-defined processing workflows

  • Prototype, develop and deploy in same framework

  • Configure and deploy to the edge

  • Deploy solutions fast!

Real-Time Video and Multi-Sensor Analytics

Easy-to-Use, Scalable, Customizable

AI Integration

Intelligent object detection, recognition and spatial tracking and customization using real-time AI processing. Quickly use both AI and traditional computer vision and processing methods.


Solutions that scale to support multiple sensor streams, mixed sensor workflows. Multiple hardware vendor support (CPU, GPU, VPU, TPU, etc.)

Multiple Sensors

Supports multiple real-time video streams, mixed resolutions and frame rates, and asynchronous workflows.


Indoor Sensing and Analytics

Quickly define and deploy solutions for retail, event centers, factories, and indoor areas that require real-time AI capabilities and solutions.

Smart City Applications

Prototype and deploy scalable solutions for smart city applications that can be customized without locking you into costly proprietary hardware or vendor specific configurations.

Security and Threat Detection

Deploy and customize real-time security solutions that provide monitoring, threat detection, tracking and real-time alerting.