Hi community,
I’m using DeepStream 6.3 with a ShopAnalytics pipeline and need to save the detected faces from the video stream. How can I do this?
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)
- DeepStream SDK Version: 6.3.0
- CUDA Driver Version: 12.2
- TensorRT Version: 8.6
- cuDNN Version: 8.9
- Hardware: NVIDIA T4 GPU
I’m trying to crop a frame around a face using the bounding box, then save it, but I’m getting this error:
RuntimeError: get_nvds_buf_Surface: Currently we only support RGBA color Format
DeepStream SDK has provided the objects images cropping and saving sample in /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-image-meta-test
Please follow the DeepStream platform compatibilities. Installation — DeepStream documentation 6.4 documentation
Hi,
I followed the example in /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-image-meta-test then modified the tiled sink and adding nvvideoconvert and “filter”. However, the pipeline crashed, and I encountered a Segmentation fault (core dumped) error. When I ran it with GST_DEBUG, I got this output in the screenshot.
Can you provide your modified code?
Hello,
I think the issue I’m facing might be related to how I link the elements in my GStreamer pipeline.
-Original Pipeline Linking:
streammux.link(queue1)
queue1.link(pgie)
pgie.link(tracker)
tracker.link(nvanalytics)
nvanalytics.link(pgie1)
pgie1.link(sgie2)
sgie2.link(sgie3)
sgie3.link(sgie4)
sgie4.link(nvstreamdemux)
I’m encountering a segmentation fault (core dumped) when I link the elements in my GStreamer pipeline like this:
streammux.link(queue1)
queue1.link(pgie)
pgie.link(nvvidconv1)
nvvidconv1.link(filter1)
filter1.link(tracker)
tracker.link(nvanalytics)
nvanalytics.link(pgie1)
pgie1.link(sgie2)
sgie2.link(sgie3)
sgie3.link(sgie4)
sgie4.link(nvstreamdemux)
Why do you add nvvideoconvert after PGIE?
I am a beginner in DeepStream, so I am following the way the elements are linked in deepstream_python_apps/apps/deepstream-imagedata-multistream-redaction/. I added nvvideoconvert because I want the frame to be converted before it reaches the tiler_sink_pad_buffer_probe. Do you have any ideas or suggestions on how to handle this properly?
To add nvvideoconvert and convert the video to RGBA format will not impact anything. Please debug your code to find out the crash point.