frigate/docs/docs/configuration/hardware_acceleration_enrichments.md

id	title
hardware_acceleration_enrichments	Enrichments

Enrichments

Some of Frigate's enrichments can use a discrete GPU / NPU for accelerated processing.

Requirements

Object detection and enrichments (like Semantic Search, Face Recognition, and License Plate Recognition) are independent features. To use a GPU / NPU for object detection, see the Object Detectors documentation. If you want to use your GPU for any supported enrichments, you must choose the appropriate Frigate Docker image for your GPU / NPU and configure the enrichment according to its specific documentation.

• AMD

  • ROCm will automatically be detected and used for enrichments in the -rocm Frigate image.

• Intel

  • OpenVINO will automatically be detected and used for enrichments in the default Frigate image.

• Nvidia

  • Nvidia GPUs will automatically be detected and used for enrichments in the -tensorrt Frigate image.
  • Jetson devices will automatically be detected and used for enrichments in the -tensorrt-jp6 Frigate image.

• RockChip

  • RockChip NPU will automatically be detected and used for semantic search v1 and face recognition in the -rk Frigate image.

Utilizing a GPU for enrichments does not require you to use the same GPU for object detection. For example, you can run the tensorrt Docker image for enrichments and still use other dedicated hardware like a Coral or Hailo for object detection. However, one combination that is not supported is TensorRT for object detection and OpenVINO for enrichments.

:::note

A Google Coral is a TPU (Tensor Processing Unit), not a dedicated GPU (Graphics Processing Unit) and therefore does not provide any kind of acceleration for Frigate's enrichments.

:::