Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
OpenFace_Shared.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@AnilRahate
Copy link

AnilRahate commented Sep 6, 2020

Hi Jin,

I am trying run colab notebook shared by you on https://towardsdatascience.com/how-to-extract-facial-expressions-head-pose-and-gaze-from-any-youtube-video-2aa6590c2bb6
However, it is not installing (build folder is not getting created hence I get error "/bin/bash: ./OpenFace/build/bin/FaceLandmarkVidMulti: No such file or directory". It would be really helpful if you can help me with this issue as I am not very conversant with programming.

@artun3e
Copy link

artun3e commented Dec 5, 2020

s not installing (bui

did you manage to solve it ? I'm having the same issue now

@jcheong0428
Copy link
Author

jcheong0428 commented Dec 7, 2020

@artun3e @AnilRahate: Hello! I re-ran the code and all the code worked fine. make sure everything is install before moving onto the next step. It takes a while to install because it neds to reinstall cuda/dlib to work with Colab.

@artun3e
Copy link

artun3e commented Dec 8, 2020

@artun3e @AnilRahate: Hello! I re-ran the code and all the code worked fine. make sure everything is install before moving onto the next step. It takes a while to install because it neds to reinstall cuda/dlib to work with Colab.

I can run it properly now, thanks!

@skwolvie
Copy link

skwolvie commented Apr 2, 2021

where is the emotion inferred?

@jcheong0428
Copy link
Author

jcheong0428 commented Apr 2, 2021

where is the emotion inferred?

@skwolvie OpenFace does not provide emotion labels, but only the action units predictions. Emotions can be inferred using a combination of AUs, for example AU12 (smiling) would be happy and AU4 (inner brow raiser) would be sad.

Alternatively, you can also use Py-Feat (https://github.com/cosanlab/py-feat) which is our new toolbox that does provide emotion predictions and should be easier to install. Here is a link to the documentation and tutorials https://py-feat.org/content/intro.html.

@rohan-kadkol
Copy link

rohan-kadkol commented Oct 1, 2021

Can OpenFace be used for real-time gaze detection? Eg. using a webcam video stream instead of a YouTube video?

@AwaisAE7
Copy link

AwaisAE7 commented Mar 8, 2022

Can i run this code on windows instead of linux?

@jcheong0428
Copy link
Author

jcheong0428 commented Mar 8, 2022

Can i run this code on windows instead of linux?

Installing and bash commands won’t work as intended as they are in linux. Nevertheless the openface github repo has window binaries you could download which actually makes installing a breeze. https://github.com/TadasBaltrusaitis/OpenFace/wiki/Windows-Installation Running the code would be similar. They python code would work either way

@jcheong0428
Copy link
Author

jcheong0428 commented Mar 8, 2022

Can OpenFace be used for real-time gaze detection? Eg. using a webcam video stream instead of a YouTube video?

Yes it’s possible. See Command line arguments in the openface wiki and search for webcam. https://github.com/TadasBaltrusaitis/OpenFace/wiki/Command-line-arguments

@AwaisAE7
Copy link

AwaisAE7 commented Mar 9, 2022

is it possible to find confident level of a person in a video using this code?

@jcheong0428
Copy link
Author

jcheong0428 commented Mar 9, 2022

is it possible to find confident level of a person in a video using this code?
@AwaisAE7 Nope, but you can train your own model from the extracted data if you have labeled data.

@Ching87
Copy link

Ching87 commented Apr 11, 2022

how did u get the processed/video.csv file?

@weskhoo
Copy link

weskhoo commented May 6, 2022

Hello jcheong0428, thank you so much for this google colab.

The google colab was working fine until yesterday where I started having this error (/bin/bash: ./OpenFace/build/bin/FaceLandmarkVidMulti: No such file or directory). Any ideas what is wrong? Thank you!

@jcheong0428
Copy link
Author

jcheong0428 commented May 9, 2022

Hello jcheong0428, thank you so much for this google colab.

The google colab was working fine until yesterday where I started having this error (/bin/bash: ./OpenFace/build/bin/FaceLandmarkVidMulti: No such file or directory). Any ideas what is wrong? Thank you!

@weskhoo Can you see a directory named "OpenFace" after running the first few cells for installation? If not, the error could be due to a failed installation. Please try running the first few cells to install OpenFace again and let me know if you still run into issues. Thanks!

@weskhoo
Copy link

weskhoo commented May 9, 2022

@jcheong0428 I do see the "OpenFace" in the directory. Attach is the image of everything is in the directory after running the first cell.
image

I recalled when I ran this weeks ago, it took 30-40 mins for the first cell, and it always ask for an input of 'Y' or 'N'. This time, it takes around 3 mins to finish running the first cell, and it does not ask me for any inputs.

I have attached the last few lines of the output of the first cell when completely run.

Reading package lists... Done
W: GPG error: http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY A4B469963BF863CC
E: The repository 'http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 InRelease' is not signed.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.
W: Target Packages (Packages) is configured multiple times in /etc/apt/sources.list.d/nvidia-machine-learning.list:1 and /etc/apt/sources.list.d/nvidia-ml.list:1
W: Target Packages (Packages) is configured multiple times in /etc/apt/sources.list.d/nvidia-machine-learning.list:1 and /etc/apt/sources.list.d/nvidia-ml.list:1

@weskhoo
Copy link

weskhoo commented May 16, 2022

Hello @jcheong0428,
I think I fixed it for now. Your google colab is working.
I think it's an error with the video drivers. I added these 2 lines in #Get newest GCC.

Get newest GCC

!sudo apt-get update
!sudo apt-get install build-essential
!sudo apt-get install g++-8

#added 5/15/2022
!sudo apt-key del 7fa2af80
!sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/3bf863cc.pub

install python dependencies

!pip install -q youtube-dl

@jcheong0428
Copy link
Author

jcheong0428 commented May 26, 2022

@weskhoo Thank you for catching the bug and fixing it! Added to the gist!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment