Qengineering / TensorFlow_Lite_SSD_RPi_64-bits

TensorFlow Lite SSD on bare Raspberry Pi 4 with 64-bit OS at 24 FPS
https://qengineering.eu/install-ubuntu-18.04-on-raspberry-pi-4.html
BSD 3-Clause "New" or "Revised" License
41 stars 6 forks source link

Only getting 9-10 FPS rate #1

Open zainmehdi opened 4 years ago

zainmehdi commented 4 years ago

Hi, First of all thank you so much for this tutorial. I followed all your steps but I am getting framerate of only 10 FPS. Can you please specify any reasons if possible. I am using RPI4B with 4GB ram.

Qengineering commented 4 years ago

Most common reason for a low frame rate are:

  1. Improper installation of OpenCV (some flags during the build where not set/reset)
  2. Improper installation of TensorFlow Lite (used build_rpi_lib.sh instead of build_aarch64_lib.sh)
  3. No overclocking (we used ± 1925 MHz to get to the high FPS)
  4. Other time consuming threads (applications) are running simultaneous.

Hopefully, this might help you to find a solution.

zainmehdi commented 4 years ago

Thanks alot for your support. I will try to make sure all the above mentioned points are followed.

Huxwell commented 1 year ago

I know the post is old, but I've recently experienced fps drop with this code on rpi 4 from 14-18fps to 2-8fps.

The reason was that by mistake I've connected a phone charger (with usb c) instead of my laptop charger - it seems that the power supply has to have enough ampers and that if this condition is not satisfied, rpi will still run, only the cpu frequencies will be lower.