2024-08-10 10:51 PM
How to figure out which model can be deployed on the board, according to the available RAM on the board.
The ST-yolo-LC model from the ST model zoo seems to fit the board's RAM availability,
(i) is it possible to train this model for my own usecase i.e fall detection and implement the same?
(ii) And if I retrain the model for my own use, can I still use the application as it is from the AI function pack examples.
(iii) What model is st-yolo-LC, is it a particular of yolo models or a custom model?
2024-08-28 02:13 AM
Hello,
On the top readme, you can see the list of the models with the supported HW for deployment.
ST YoloLC (LC stands for Low Complexity) is a ST custom model derived from TinyYolov2. Yes, you can retrain this model with your dataset.
This model is an Object Detection (OD) model, so as soon as you are referring to an OD use case, you should be able to use the stm32ai application code as well.
Don't hesitate to communicate on the model zoo GH in case of issue.
Regards,
2024-08-28 08:34 PM
Thank you for your reply, I have used a mobilenet model with the person detect application from the FP-AI-Vision function pack, I've also modified the postprocess function such that once a fall is detected, it checks for the next 10 seconds if the person is still fallen down or has recovered. If they are still fallen down a alert message is displayed else recovered message is displayed.
But I want to take this further and connect GSM sim900a module and make alert calls if the person doesn't recover , but I'm facing problems with this.
I'm not able to setup usart communication between the stm32h747 board and GSM module despite the connections, baud rate everything being correct.
Kindly help me with this also,
And is it possible to get the ioc file for the person detect application present in the function pack?