Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

Training on Custom dataset for LPR

titir_s_Intel
Employee
462 Views

Hi, 

 

What is the minimum no. of images that I need to train LPR on my custom dataset:

https://github.com/openvinotoolkit/training_extensions/tree/develop/misc/tensorflow_toolkit/lpr

 

Since, its hard to get annotated USA license plate images, like the CCPD(chinese) dataset, a minimum no. will be helpful.

 

BR,

Titir

0 Kudos
2 Replies
IntelSupport
Moderator
430 Views

Hi Titir,

Thanks for reaching out.

There is no minimum number of the exact datasets that we can provide. However, too small number of trained images will not enough for the model to give better performance. We would recommend you use the recommended number of datasets used for the project, which are 300320 images.

 

Regards,

Aznie


0 Kudos
IntelSupport
Moderator
399 Views

Hi Titir,


This thread will no longer be monitored since we have provided a solution. If you need any additional information from Intel, please submit a new question.


Regards,

Aznie



0 Kudos
Reply