Using Rivermax SDK for processing each pixel from the video stream

Hi,

I’m completely new in Nvidia products. I would like to use Rivermax SDK to get an HDMI 1080P stream and convert it to the Network stream with pre-processing. Pre-processing will be based on JSON scenarios loaded from the user space wich will affect each pixel from the video stream.
Is it possible to get this using rivermax? And, Do I need to write a special code for the DPU processor?

Many thanks

UPD,

Hardware with RDMA support

  1. NVIDIA Mellanox MCX623106AN-CDAT ConnectX®-6 Dx EN Network Interface Card
  2. Nvidia-rtx-4000-ada

@michaelbe Maybe you can address this question?

Rivermax just use to send/receive over RDMA/SDP will not cover pixel processing.

And rivermax is licensed software need pay for license.

FYI,

1 Like

Thank you for your reply. So no any chance to create DPU routine or such for making pixel processing and create own SDP profile for sending in your own format using RDMA and NIC?

It’s not a problem to pay for the license.

Rivermax just a communicate lib, not include pixel processing. Of cause you can use other libs base on cpu/gpu do that, then use rivermax to transmit.

1 Like

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