I cannot get the unchanged gen2-license-plate-recognition script to recognise license plates in either the video or camera options. The code runs and detects the vehicles with bounding boxes and reports vehicle attributes in a separate window but the license plate is not detected or reported. With "python main.py -vid ./chinese_traffic.mp4" the Win11 Command window displays:
Starting pipeline...
Creating pipeline...
Creating License Plates Detection Neural Network...
Creating Vehicle Detection Neural Network...
Pipeline created.
[18443010F107C80F00] [1.2] [1.395] [DetectionNetwork(0)] [warning] Input image (300x300) does not match NN
(3x300)
The last line with the Detection Network warning is continuously repeated while the code runs.
I have an OAK-D Lite and I am using Python 3.12
The installed package versions are:
depthai 2.29.0.0
depthai-sdk 1.15.1
opencv-python 4.10.0.84
blobconverter 1.4.3
I tried using blobconverter 1.2.8 and it didn't run at all, reporting several incompatabilities.
Could one of the package versions be the problem or is something else wrong?
Any help would be greatly appreciated as I would like to use a derivative of this example in a project I am hoping to progress.
