I’m working on a detection (pgie) - alignment (image processing) - recognition(sgie) pipeline. I want to use the output from detection (key points) and full frame to do alignment before feeding the aligned image to sgie. The point here is keeping the whole pipeline on gpu. Can you give me some suggestions?
Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
For your question, I think you can do that on the sgie
Idon’t know how I can do image processing (with opencv) on sgie. My intention is using customized gst-dsexample. For now I’m having trouble with getting the frame from buffer.
In addition, I don’t understand why nvidia uses cpu version of opencv for processing there. Isn’t it a bottleneck in the pipeline?