Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU): GPU
• DeepStream Version: 6.0
**• TensorRT Version: 8.2 **
• NVIDIA GPU Driver Version (valid for GPU only): 470.82
• Issue Type( questions, new requirements, bugs): question
I have this line to run the basic example but cropping the video source before inference using nvvideoconvert, which works perfect!
gst-launch-1.0 filesrc location=/opt/nvidia/deepstream/deepstream-6.0/samples/streams/sample_1080p_h264.mp4 ! decodebin ! m.sink_0 nvstreammux name=m batch-size=1 width=1280 height=720 ! nvvideoconvert src-crop=100:100:500:400 ! nvinfer config-file-path=/opt/nvidia/deepstream/deepstream-6.0/samples/configs/deepstream-app/config_infer_primary.txt ! “video/x-raw(memory:NVMM),width=500,height=400,format=RGBA” ! nvdsosd ! nveglglessink
However, what I’m tring to do is crop the image inside the config file, let’s say “deepstream_app_config_yoloV3.txt” (or anyone), if i try to follow the same order my file looks like this (default file):
[application]
…
[tiled-display]
…
[source0]
…
[sink0]
…
[osd]
…
[streammux]
…
[nvvideoconvert]
src-crop=100:100:500:400
[primary-gie]
…
[tracker]
…
I also tried renaming to [videoconvert],[nvdsvideoconvert],[video-convert] but it didn’t work either
I read other posts but I couldn’t find someone who did. Is it possible to do this from the config file?