Introduction
Media Server can use a graphics card (GPU) to perform some processing tasks:
-
Analysis tasks. Using a GPU rather than the CPU can significantly increase the speed of training and analysis tasks that use Convolutional Neural Networks. The analysis tasks that benefit from GPU acceleration are:
- Face detection, recognition, and demographics.
- Speech to text.
- Image classification.
- Object class recognition.
- Vehicle make recognition.
- Optical Character Recognition (video ingest and scene mode only).
- Ingesting (decoding) and encoding video. You can use the GPU to decode and encode some types of video. This can reduce the load on the CPU, releasing resources for media analysis.