dlib and face_recognition on raspberry pi.md

Install dlib and face_recognition on a Raspberry Pi

Instructions tested with a Raspberry Pi 2 with an 8GB memory card. Probably also works fine on a Raspberry Pi 3.

Steps

Download the latest Raspbian Jessie Light image. Earlier versions of Raspbian won't work.

Write it to a memory card using Etcher, put the memory card in the RPi and boot it up.

Log in. Default username / password is pi / raspberry.

Set up Wifi (if you are using Wifi) according to the Raspberry Pi instructions.

Run sudo raspi-config and configure the basics:

• Set up your keyboard layout (It defaults to a British keyboard layout)
• Change default user password
• Enable the Raspberry Pi camera (if you have one attached)
• Configure gpu memory split under 'Advanced'. Set it up '16'.
• Save changes and reboot.

Install required libraries with these commands:

sudo apt-get update
sudo apt-get upgrade
sudo apt-get install build-essential \
    cmake \
    gfortran \
    git \
    wget \
    curl \
    graphicsmagick \
    libgraphicsmagick1-dev \
    libatlas-dev \
    libavcodec-dev \
    libavformat-dev \
    libboost-all-dev \
    libgtk2.0-dev \
    libjpeg-dev \
    liblapack-dev \
    libswscale-dev \
    pkg-config \
    python3-dev \
    python3-numpy \
    python3-pip \
    zip
sudo apt-get clean

Install the picamera python library with array support (if you are using a camera):

sudo apt-get install python3-picamera
sudo pip3 install --upgrade picamera[array]

Temporarily enable a larger swap file size (so the dlib compile won't fail due to limited memory):

sudo nano /etc/dphys-swapfile

< change CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024 and save / exit nano >

sudo /etc/init.d/dphys-swapfile restart

Download and install dlib v19.6:

mkdir -p dlib
git clone -b 'v19.6' --single-branch https://github.com/davisking/dlib.git dlib/
cd ./dlib
sudo python3 setup.py install --compiler-flags "-mfpu=neon"

Install face_recognition:

sudo pip3 install face_recognition

Revert the swap file size change now that dlib is installed:

sudo nano /etc/dphys-swapfile

< change CONF_SWAPSIZE=1024 to CONF_SWAPSIZE=100 and save / exit nano >

sudo /etc/init.d/dphys-swapfile restart

Download the face recognition code examples:

git clone --single-branch https://github.com/ageitgey/face_recognition.git
cd ./face_recognition/examples
python3 facerec_on_raspberry_pi.py

Totally Optional: If you want a desktop GUI, install PIXEL:

sudo apt-get install --no-install-recommends xserver-xorg xinit raspberrypi-ui-mods

Hello, is it still relevant in 2024?
I have Raspberry Pi 4 Model B 4G or Raspberry Pi 5 8GB with latest OS (Bookworm).
I tried at least 3 or 4 other alternatives and I'm sick of issues and errors. Did it anyone made it lately?
I will try today on Raspberry Pi4 Model B 4G.

The library (face_recognition) continues working well for me on a Raspberry Pi 5 with the Raspbian Bookworm OS. However, all steps change due to various changes in the very OS. As I compiled OpenCV first, I changed and prepared many system packages, and prevented me further headaches. After that, I was able to install dlib and face recognition using pip, both at the most recent versions.

The library (face_recognition) continues working well for me on a Raspberry Pi 5 with the Raspbian Bookworm OS. However, all steps change due to various changes in the very OS. As I compiled OpenCV first, I changed and prepared many system packages, and prevented me further headaches. After that, I was able to install dlib and face recognition using pip, both at the most recent versions.

Hi am currently working on a Raspberry Pi 4 with Raspbian Bookworm OS. I am struggling with finding the correct steps to download dlib and face recognition (I successfully downloaded OpenCV). Can you please give me a quick rundown on how you downloaded dlib and face recognition because whenever I try installing them I keep get an error about an "externally managed environment." Any help or guidance would be greatly appreciated! Thank you!!

Hello, What steps were changed, please describe.

…

On 20.06.2024 17:52, mardelmariam wrote: Re: ageitgey/dlib and face_recognition on raspberry pi.md ***@***.**** commented on this gist. ------------------------------------------------------------------------ The library (face recognition) continues working well for me on a Raspberry Pi 5 with the Raspbian Bookworm OS. However, the steps change a lot, because I compiled OpenCV first. The compilation changed many packages and prevented me further headaches. After that, I was able to install dlib and face recognition, both at the most recent versions. — Reply to this email directly, view it on GitHub <https://gist.github.com/ageitgey/1ac8dbe8572f3f533df6269dab35df65#gistcomment-5095302> or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABJ5OROCUCAGMWZXLPBJ4YLZIL3DBBFKMF2HI4TJMJ2XIZLTSKBKK5TBNR2WLJDUOJ2WLJDOMFWWLO3UNBZGKYLEL5YGC4TUNFRWS4DBNZ2F6YLDORUXM2LUPGBKK5TBNR2WLJDHNFZXJJDOMFWWLK3UNBZGKYLEL52HS4DFVRZXKYTKMVRXIX3UPFYGLK2HNFZXIQ3PNVWWK3TUUZ2G64DJMNZZDAVEOR4XAZNEM5UXG5FFOZQWY5LFVA2DMOBVG4ZTEMFHORZGSZ3HMVZKMY3SMVQXIZI>. You are receiving this email because you commented on the thread. Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.

Hello, This i find interesting. Have you any literature for me? Many thanks

…

On 22.06.2024 08:32, OmarAKhan1 wrote: Re: ageitgey/dlib and face_recognition on raspberry pi.md ***@***.**** commented on this gist. ------------------------------------------------------------------------ The library (face_recognition) continues working well for me on a Raspberry Pi 5 with the Raspbian Bookworm OS. However, all steps change due to various changes in the very OS. As I compiled OpenCV first, I changed and prepared many system packages, and prevented me further headaches. After that, I was able to install dlib and face recognition using pip, both at the most recent versions. Hi am currently working on a Raspberry Pi 4 with Raspbian Bookworm OS. I am struggling with finding the correct steps to download dlib and face recognition (I successfully downloaded OpenCV). Can you please give me a quick rundown on how you downloaded dlib and face recognition because whenever I try installing them I keep get an error about an "externally managed environment." Any help or guidance would be greatly appreciated! Thank you!! — Reply to this email directly, view it on GitHub <https://gist.github.com/ageitgey/1ac8dbe8572f3f533df6269dab35df65#gistcomment-5097037> or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABJ5ORJDRHWAHWNJKGKSA7TZIULBJBFKMF2HI4TJMJ2XIZLTSKBKK5TBNR2WLJDUOJ2WLJDOMFWWLO3UNBZGKYLEL5YGC4TUNFRWS4DBNZ2F6YLDORUXM2LUPGBKK5TBNR2WLJDHNFZXJJDOMFWWLK3UNBZGKYLEL52HS4DFVRZXKYTKMVRXIX3UPFYGLK2HNFZXIQ3PNVWWK3TUUZ2G64DJMNZZDAVEOR4XAZNEM5UXG5FFOZQWY5LFVA2DMOBVG4ZTEMFHORZGSZ3HMVZKMY3SMVQXIZI>. You are receiving this email because you commented on the thread. Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.

Anyone have any luck installing face_recognition on a Raspberry Pi 5 (running Bookworm)? All the instructions on the internet are out of date...

For anyone who is struggling with the installation of the required libraries: you will need to compile OpenCV, so the OS will run C/C++ files instead of running OpenCV with the Python files. A guide for this step is here. If some errors appear, Google is your friend. After installing OpenCV, you may install dlib, and that should work.

Hello, if you need to install face_recognition, you will need to compile OpenCV first: Install OpenCV on Raspberry Pi 5 - Q-engineering (qengineering.eu) <https://qengineering.eu/install%20opencv%20on%20raspberry%20pi%205.html> After that, you may install dlib, and then, face_recognition. I don't remember the steps. However, I remember that this process of making this library work isn't too difficult. Wish you success on the task! El mié, 11 sept 2024 a las 16:27, Cameron Smith ***@***.***>) escribió:

…

***@***.**** commented on this gist. ------------------------------ Anyone have any luck installing face_recognition on a Raspberry Pi 5 (running Bookworm)? All the instructions on the internet are out of date... — Reply to this email directly, view it on GitHub <https://gist.github.com/ageitgey/1ac8dbe8572f3f533df6269dab35df65#gistcomment-5186507> or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANPZVDFONL34DW6VDBX5QW3ZWCYT5BFKMF2HI4TJMJ2XIZLTSKBKK5TBNR2WLJDUOJ2WLJDOMFWWLO3UNBZGKYLEL5YGC4TUNFRWS4DBNZ2F6YLDORUXM2LUPGBKK5TBNR2WLJDHNFZXJJDOMFWWLK3UNBZGKYLEL52HS4DFVRZXKYTKMVRXIX3UPFYGLK2HNFZXIQ3PNVWWK3TUUZ2G64DJMNZZDAVEOR4XAZNEM5UXG5FFOZQWY5LFVA2DMOBVG4ZTEMFHORZGSZ3HMVZKMY3SMVQXIZI> . You are receiving this email because you commented on the thread. Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub> .

-- ~ Maryam del Mar Correa ~

so this is how it worked for me on raspberry pi 3b
i did zram, increased swap file to 2gb, stopped the desktop environment (lightdm) and other services that i don't need and booted into console to save as much ram as possible (i got it to 95mb).
i installed dependencies of opencv:
sudo apt-get install -y build-essential cmake git unzip pkg-config libjpeg-dev libpng-dev libavcodec-dev libavformat-dev libswscale-dev libgtk2.0-dev libcanberra-gtk* libgtk-3-dev libgstreamer1.0-dev gstreamer1.0-gtk3 libgstreamer-plugins-base1.0-dev gstreamer1.0-gl gstreamer1.0-libcamera libxvidcore-dev libx264-dev python3-dev python3-numpy python3-pip libtbbmalloc2 libtbb-dev libv4l-dev v4l-utils libopenblas-dev libatlas-base-dev libblas-dev liblapack-dev gfortran libhdf5-dev libprotobuf-dev libgoogle-glog-dev libgflags-dev protobuf-compiler
and then:
pip install opencv-python pip install imutils
and then:
pip install --no-cache-dir dlib
i took around 3 hours to install but it was successful (try to save as much ram as possible for it)
and then:
'pip install face-recognition'

Hey @LouayTselkim,

Thanks for sharing! That is great!
Can you share the OS version as well? I m on Trixie and am stuck at "libatlas-base-dev". It is not there anymore.
Seems this support it.
I don`t know which is better? Downgrade OS version or try to compile it from source code.