How to setting the different labels.txt while using two models in the same program


TensorRT Version :
CUDA Version :
Operating System + Version :
Jetson NX (jetpack 4.6)

I want to use the detection and number recognition from this repo (GitHub - winter2897/Real-time-Auto-License-Plate-Recognition-with-Jetson-Nano: This repository provides you with a detailed guide on how to build a real-time license plate detection and recognition system. The source code of the repository implemented on Jetson Nano reached 40 FPS.) in same program, however I have no idea on setting 2 different class_labels since there cannot find the argument setting about labels in the code but need to set in command line like

python3 --model=./networks/az_plate/az_plate_ssdmobilenetv1.onnx --class_labels=./networks/az_plate/labels.txt --input_blob=input_0 --output_cvg=scores --output_bbox=boxes --overlay=none images/car.jpg out.jpg

Any idea on this?

Also how to set 2 different models, since the --model can only be set in command line not in code…


Have you checked this with the repository owner?

It looks like the sample only supports a single detection model.
But you can add it easily by modifying the source below:


I am confusing that why I set --network=./network/az_plate/az_plate/az_plate_ssdmobilenetv1.onnx will get differenct detection result from setting --model=./network/az_plate/az_plate/az_plate_ssdmobilenetv1.onnx?

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.