.Limit on number of faces registered

[7thstorm]

Can someone explain why there is a limit on the number of faces that can be registered per person?
I started with 250 images and it gave me errors

    dsm.py -H localhost:777 register -p "../faces/east/" person1
{'success': False, 'error': 'error occured on the server', 'duration': 0}

However, after trying various different number of images, it seems like 25 is the maximum.

is there a way to increase this limit? In my use case, I am using some augmented images, so need to be able to use 100 or more. This is very frustrating.
Thank you!

[sickidolderivative]

I faced the same issue few months ago on my 4GB Jetson Nano. Submitting more than 25 pictures would make the device completely unresponsive. It seemed the limit was related to Nano’s limited resources. Please take a look on the following thread with valuable input from John and Robin:

DeepStack Release on Jetson

Happy to report that I installed deepstack on my 4GB jetson nano. Object detection seems to work fine, with average processing time of 130 ms, or 7 fps consuming 30% and 60% of my nano’s CPU and memory respectively. Face recognition works partially. I was successful training one face with 10 photos, but 100 or 250 photos seems to hang deepstack infinitely. I waited over an hour registering 250 photos (something that takes 2 minutes 30 seconds on RPI+NCS2) with no luck. During the hour of waitin…

Having moved to 8GB Jetson Xavier, I’ve been able to submit up to 80 photos in one request without freezing the device. Perhaps that’s something worth considering in your use case.

[7thstorm]

I’m not sure if this is the issue. My system has a 6GB GPU and 32GB of RAM
The error I received was within 5 seconds
I started at 250 images, then tried 100, then 50, 30 and it worked at 25

I’ll try again

[7thstorm]

It does not work for 26 images, but will work for 25