Skip to content

Using Tensor Flow and Machine learning to teach computer both image and live object detection. Currently it can recognizes facial expressions and gestures.

Notifications You must be signed in to change notification settings

unstoppablo/Live-Object-Recognition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tensor-Flow-Object-Recognition

___________________________________________________________

Model has been trained to recognize a variety of things, including expressions such as smiles, frowns, squints, and more.

Adapted from Nicholas Renotte's 5-hour Youtube course on Tensor Flow object recognition. Course focused on teaching hand gestures, adapted it to recognize facial expressions.

Here is an example of facial expression recognition: Screen Shot 2021-08-16 at 3 58 25 PM

___________________________________________________________

Errors I encountered and fixes:

Error: ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject Solution: Reinstall pycocotools to a matching version for Python and Tensor Flow. Solution 2: Reinstall numpy or try changing the versions to match the required architecture.

Helpful links that helped me fix this issue: https://stackoverflow.com/questions/66060487/valueerror-numpy-ndarray-size-changed-may-indicate-binary-incompatibility-exp scikit-image/scikit-image#5270 scikit-learn-contrib/hdbscan#457

Error: ValueError: 'images' must have either 3 or 4 dimensions. Solution: This happened with image detection, usually the image name and path are incorrect. For example, if in section 9 you are giving it the path for the image you want it to recognize, but the filetype is "png" and you give it another one, this error will appear.

___________________________________________________________

Resources: https://www.youtube.com/watch?v=yqkISICHH-U&ab_channel=NicholasRenotte

Coding Requirements:

  • Python 3.8 or higher
  • Tensor Flow --upgrade (2.3 currently)
  • Protobuf matplotlib==3.2
  • Numpy 1.20.0

Note: This project was built within a virtual environment. Bellow are the required dependencies and commands to install them

  • python -m pip install --upgrade pip
  • pip install ipykernel
  • python -m ipykernel install --user --name=YOUR ENVIRONMENT NAME

About

Using Tensor Flow and Machine learning to teach computer both image and live object detection. Currently it can recognizes facial expressions and gestures.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published