Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) dGPU
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only) 510
• Issue Type( questions, new requirements, bugs) question
• 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)
Hello I am trying to create a deepstream pipeline that has a TEE and outputs to 3 branches.
# ,-> vidconv1 -> jpegenc1 -> multifilesink1
# nvinfer -> capsfilter -> tee -> demux -> vidconv2 -> jpegenc2 -> multifilesink2
# `-> appsink
# `-> queue -> tiler -> dsosd -> eglsink
- nvstreamdemux,
- appsink and
- tiler
All my application logic is in appsink. It was working fine without the tiler. But after I added the tiler, it transforms nvdsbatchmeta and the detection bbox coordinates are all transformed by the tiler. So a bbox in src-id=2, is moved by more than 1 feed width depending on how the tiler calculated its position. My application logic fails due to this messup.
Is there a way for me to keep the tiler branch’s copy of nvdsbatchmeta isolated from appsink’s copy of nvdsbatchmeta? I think even a simple copy plugin can help with that… or can I push to tiler after appsink?
Is there a gstreamer/deepstream plugin that will let me do that?
How to achieve this? Are there any alternative ideas/suggestions? Please help!