riya-17 / facerecognition Goto Github PK

Rotate the images to 90 or 180 degree angles.

I am getting this error while running FaceRecognizer.py

Traceback (most recent call last):
  File "FaceRecognizer.py", line 67, in <module>
    predictor = dlib.shape_predictor(args["shape_predictor"])
RuntimeError: Unable to open complete_path/shape_predictor_68_face_landmarks.dat

Although I followed the steps to build and install dlib and I guess it is installed successfully.

>>> import dlib
>>> print(dlib.__version__)
19.17.0

what am i missing?

Hello, I found several ways to implement emotion recognition to the project.

1. Go with Emotion API by Microsoft it is free with following limitations

I believe that it is suitable for non-real-time recognition task.
2. Apply trained predictive models from existing open source project as Emotion recognition with CNN, Emotion or face classification. All of them are trained on same dataset and based on same article
3. Use deep learning approaches to train our own model for the task.

@riya-17 , what do you think we should go with?
I can start to work on implementation by opening a PR.

I am using an Windows system and I have a hard time to set this up. The parts of I figure out are Prerequisites and Setup. So, can I do a Pull request about these in README.md? @riya-17

Enhance rotateImage.py

In main function, file path is not optimized for all users like this.

image = "C:\\Users\\KSHITIJ\\PycharmProjects\\rotateImage\\virat.jpg"

Programming Language

• Python

I am using an Ubuntu system and I am having a hard time setting this up. If I figure it out I will try and do a Pull request

Hi, @riya-17 !
I would like to contribute to the project!
The proposition is to wrap all required installation steps to requirements.txt and call pip just once as follows: pip install -r requirements.txt
What do you think about it?
I could add this file and fix description on main page.

We can improve this project by adding new features to the current project like:

• rotating the image
• Detecting emotions of the image
• zoom-in or zoom-out
  and many others

I got that error when i do pip install. I think it must be opencv-python instead of opencv. I'll send a PR for it.

I think that the main logic of FaceRecognizer would be most useful if it was separated into functions that perform the important tasks with an optional main function that runs the program as it currently is.

This would make the code more reusable for expanding further.

This project still needs a lot of functionalities like :

• Rotation of images
• Detect Emotions
• Zoom-in or zoom-out

Hello @riya-17,
Can I contribute to this project by adding new features to the current project like:
Emotion Recognition
Improved Accuracy (If we use neural networks for the same)
Creating a web interface for the same and deploying the created machine learning model

We can improve the same by adding more images to the database.

Would it be fine if I work on this?

increase or decrease the size of image

There's a problem in installing setup.py
Since cmake compiler is not set and neither specified in prerequisites, thus the command (cd ..) is throwing an error. The error statement is given below.

CMake Error at CMakeLists.txt:3 (project):
Running

'nmake' '-?'

failed with:

The system cannot find the file specified

CMake Error: CMAKE_C_COMPILER not set, after EnableLanguage
CMake Error: CMAKE_CXX_COMPILER not set, after EnableLanguage
-- Configuring incomplete, errors occurred!

This can be resolved by giving a procedure/command to set cmake compiler in the prerequisites block