Deepstream-lpr-app not running with in_video


this is a reference topic to : Deep stream 5.0 LPR application not compiling - #6 by yash.j

The issue with the gst-streammux is causing the issue in running the deepstream-lpr-app
Any help on how to test the license plate detection with proper input video to enter would be of great help.


In deepstream_lpr_app/ at master · NVIDIA-AI-IOT/deepstream_lpr_app · GitHub, the command line has been shown.

./deepstream-lpr-app <1:US car plate model|2: Chinese car plate model> \
         <1: output as h264 file| 2:fakesink 3:display output> <0:ROI disable|1:ROI enable> \
         <input mp4 file name> ... <input mp4 file name> <output file name>

Your command line is wrong. The output file is missed.

Hi @Fiona.Chen

Thanks for the response.
I have used command as :
./deepstream-lpr-app 1 3 0 input.mp4 output.mp4

where input.mp4 is a video of cars moving in traffic, output.mp4 is file I expect to be created having required LPR detection.

Kindly suggest pre-tested input/output files if any.

What do you want?

How many input files should we include ?
And what should be the content in it ?

The models used in this application is to recognise car license characters. Seems you have choose to identify Chinese car license, so you can take any video where there are cars with Chinese licenses.

At most 128 input videos is allowed with this sample app.