Versions of stm32 ecosystem for NUCLEO-N657X0-Q

[whoami]

Hello ST community!

Currently i'm working on a project based on the stm32 NUCLEO-N657X0-Q which should be delivered very soon .
I followed this article https://community.st.com/t5/stm32-mcus/how-to-build-an-ai-application-from-scratch-on-the-nucleo-n657x0/ta-p/828502 step by step with the same versions of software mentioned but without any results .

I tried using X-Cube AI and the new tool Stm32Cube AI studio but i ended up with a lot of confusion and errors .

I'm planning to stick for now with x-cube ai since the new tool has a lot of bugs to be fixed .

My questions is what are the right versions of the software to use to get a computer vision (Yolo like) to work on my board ?

Thnx!

[hamitiya · Accepted solution]

Hello @whoami 

Could you describe more what you mean by "a lot of bugs". For NUCLEO-N657X0-Q, I am not aware of any issues with the latest version of STM32Cube AI Studio.

For the S/W version, if you want to work with X-CUBE-AI (10.2.0), you will need

- ST Edge AI Core version 2.2.0

- STM32CubePrg 2.21

- STM32CubeIDE 2.1.0

You can also use models and deployment scripts from ST Model Zoo.

Best regards,

Yanis

[whoami]

Hello @hamitiya !

Thank you for reply!

I think the problem is with CubeMX as the community mention.

I used CubeMX 6.16.0 with CubeIDE 2.1.0 and St edge ai 4.0 and i managed to work and the model is giving correct results.

Now I'm struggling with interfacing an STEVAL- 66GYM camera within my project.

I don't have a screen to plot the results on it , just want to store the camera data into memory array will be then passed to a NN model.

Is there any documentation or article to follow for the NUCLEO-N657X0-Q .

Thanks!

[Julian E.]

Hi @whoami,

We are glad to hear that you solved your first issue.

Regarding how to use a camera, we don't have a simple tutorial to do it but we have a lot of getting started / package that most likely achieve what you are trying to do.

STM32N6-AI | Software - STMicroelectronics

I would suggest looking at the image classification getting started, which should be the most straightforward one and probably the "x-cube-n6-camera-capture" that I don't know, but seems to be linked to your question.

Have a good day,

Julian

[whoami]

Hi @Julian E.  ! 
Thank you for your reply !

I consulted this repo https://github.com/STMicroelectronics/x-cube-n6-camera-capture/tree/main and using the binary i managed the camera to work as a webcame , but now i'm trying to build my own project but im finding a problems getting it to work .
I'm trying to find  what configuration should be done through reverse engineering the code of this project. However since this project implements a lot other topics (FreeRTOS , usb ...) im getting stuck within this code .

For now i managed i figured that i should use the pipe 1 for my project but still not getting how to configure these parameters .

 

 I had also read this application note AN6211 but still not getting it right 
is ther any simple project that i can consult !

Thnx!

[Julian E.]

Hi @whoami,

I would suggest looking at the cubeN6firmware and the example in it to try to find a simple code for Image Capture. In

STM32Cube_FW_N6_V1.3.0\Projects\STM32N6570-DK\Examples I see a DCMIPP and JPEG examples, they probably contain some help.

For the AI part, I would suggest to use CubeAI Studio with your model and generate the code. You will have a simple application that just do a AI inference with random data.

Have a good day,

Julian