i want to train https://github.com/opencv/openvino_training_extensions/tree/develop/tensorflow_toolkit/lpr with my custom dataset, i checked Synthetic Chinese License Plates and all these images have single line text, but my custom data for indian number plate also have two line number plates, like below
can i train this lpr for this type of license plates.
If you had followed this official Openvino tutorial: https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_installing_openvino_windows.htm...
You will end up running a model that detect a car and the plate number is also identified.
The Openvino open zoo models also already have pre-trained models for these kind of purposes.
Training your own model based on this should have no problems even for two lines number plate. The rest is depend on how you extract the data.
Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question