Skip to main content
Version: 2.4.0 (latest)

Human Activity Recognition (HAR)

Human Activity Recognition (HAR) include recognition of human falls, fights, lying and sitting in video streams and video files.

System Requirements

The following configuration is required to enable HAR on a single Full HD @ 25fps video stream:

Minimum system requirements +

  • GPU: performance ~FP32 10 TFLOPS; 3GB VRAM; CUDA11 Runtime;
  • RAM: +4Gb;
  • CPU: 2 AVX cores @3GHz.

When an activity type is detected, OMNI Agent generates an activity process and sends it to OMNI Platform or to a third-party service.

How to Enable HAR

Open the run_params.json configuration file and set "true" in the pipelines.action_recognition.enabled field.


When enabling HAR, we recommend using:

  • CUDA: set "true" in the use_cuda_onnx field of the run_params.json file or enable "CUDA usage" option in Settings of OMNI Agent web-interface.

For stable work of OMNI Agent, disable the generation of face biometric templates. To do this, open the run_params.json configuration file and set "false" in the pipelines.face.enable_templates_generation field.