Documentation for Axxon Next 4.6.0. Documentation for other versions of Axxon Next is available too.

Previous page Fine-tuning the Face detection tool  Configuring Face detection (VL) Next page

Starting from DetectorPack 3.8, a utility was added to the Face Recognition Pack add-on (see Installing DetectorPack addons), which allows you to create GPU network caches without using Axxon Next. The presence of the cache speeds up the initialization and optimizes video memory consumption.

To optimize the operation of the Face detection on GPU, do the following:

  1. Stop the Server (see Shutting down a Server).

    Attention!

    If the system has the software running on GPU, it is necessary to stop its operation. 

  2. Create the GPU_CACHE_DIR system variable (see Appendix 10. Creating system variable) by specifying in the Variable value field the path to the cache location with an arbitrary folder name. For example, D:\AN_GPU_cache. The specified directory will store the cache for all used detection tools and neural networks.

  3. To call the utility, open in the command line: C:\Program Files\Common Files\AxxonSoft\DetectorPack\TvaFaceGpuCacheGenerator.exe (1) and press Enter.


  4. Specify the ID of the required GPU (2, see General Information on Configuring Detection) and press Enter.

  5. If you want to change the resolution of the video stream for the Face detection, enter y (3) and specify the required resolution (4).

    Note

    GPU uses the default video stream resolutions. Changing the resolution of the video stream will increase the percentage of face detection when using the resolutions from 1920*1080 and higher.

  6.  Press Enter. The cache creation process will start. It will take about 20 minutes. The cache creation was successful, if the result value is “0”. 

Optimizing the operation of Face detection on GPU is complete.

Attention!

If you update the Face Recognition Pack add-on (see Installing DetectorPack addons) or change the NVIDIA GPU model, the cache must be recreated.

  • No labels