Deepstream processing in custom detector takes place on CPU or GPU?

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson Nano
• DeepStream Version 5.1
• JetPack Version (valid for Jetson only) 4.5
• TensorRT Version 7.1
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs) Question

Does the Post processing in custom function parser implemented in Deepstream take place on GPU or is it CPU?
I have implemented custom NvDsInferParseCustom function and want to know if it is not processed on CPU then is it possible to process some part of it on GPU for faster computation?

hi @mehul.vast
The NvDsInferParseCustom function can be passed in the input plugin nvinfer and the nvinfer works on the hardware support GPU/DLA/VIC on Jetson. You can refer these documentation for more understanding it will give you a clear idea. NVIDIA DeepStream SDK API Reference: nvdsinfer_custom_impl.h File Reference.
Gst-nvinfer — DeepStream 5.1 Release documentation

Thank you

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.