Hello All,
I'm currently working on IMX8MM Board for an eIQ ML project.
Here I have successfully executed the TensorFlow Demo and its working fine.
$ cd /usr/bin/tensorflow-lite-2.12.1/examples
$ ./label_image -m mobilenet_v1_1.0_224_quant.tflite -i grace_hopper.bmp -l labels.txt
But while running the face_recognition with the live camera. I've been facing the below attached error.
$ cd /usr/bin/eiq-examples-git/face_recognition
$ python3 main.py -i /dev/video0
Traceback (most recent call last):
File "/media/eiq-example-lf-6.1.1_1.0.0/face_recognition/main.py", line 29, in <module>
detector = YoloFace("../vela_models/yoloface_int8_vela.tflite", args.delegate)
File "/media/eiq-example-lf-6.1.1_1.0.0/face_recognition/face_detection.py", line 21, in __init__
self.interpreter = tflite.Interpreter(model_path=model_path)
File "/usr/lib/python3.9/site-packages/tflite_runtime/interpreter.py", line 365, in __init__
_interpreter_wrapper.CreateWrapperFromFile(
ValueError: Could not open '../vela_models/yoloface_int8_vela.tflite'.
We are following the below attached Document.
https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf
But in the document it have mentioned that it will work on the imx93 board. But I want the image recognition software to run on imx8mm and imx8mp boards. Can you please help me with a solution.
Thanks in advance.
Hi @mohamed_gaseen,
Thank you for contacting NXP Support.
The examples included on the path 9 eIQ Demos are specifically for the i.MX93 NPU. Therefore, you can't run these demos on your targets iMX8MM and iMX8MP.
You can have a look to the GoPoint application demo for i.MX Application processors.
GoPoint for i.MX Applications Processors User Guide (nxp.com)
I hope this information will be helpful.
Have a great day!