Neucrack

`img.pix_to_ai()` is copy image content to kpu input buffer, if the image from `sensor.snapshot()`, it already have, if not, e.g. read from file, or some image ops like copy resize...

k210 can do this by fpioa, but we not implement this on maixpy

maybe some error occured when boot up, connect to PC, open terminal tool to see boot up log to find error reason

CI check failed 

actually this func is not full tested, it's better to use model to find face https://github.com/sipeed/MaixPy_scripts/blob/master/machine_vision/demo_find_face.py

try latest firmware

what's the phenomenon and code of recurrence?

@Yabuku0 you can find train code here: https://github.com/sipeed/maix_train and docs here: https://wiki.sipeed.com/soft/maixpy/zh/index.html , and you can use translate 

you can point out which API you need have bug, we can fix it

you can manually compile MaixPy project, and select modules you need, for example video:  video module not occupy too much memory, there's the compare of include video or not...