Skip to content

Gender Prediction using Deep Learning Algorithms and Model on Images of an Individual.

License

Notifications You must be signed in to change notification settings

vilas897/Gender-Detection

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Gender-Detection

Dataset

Data Preprocessing

  • Few of the Images were corrupted and hence had to be removed from the dataset.
  • Face Detection Algorithm (Cascade Classifier) is used to check if an image is valid or not.
  • If a face is detected in the Image, Collect the required details (Full path and Gender) of the Image.
  • Store the required data in a different .csv file.

Framework/Library used

  • Keras with Theano backend.
  • OpenCV2 for Image Processing Algorithms.
  • Pandas and NumPy for .csv and mathematical computations.

Keras

  • Is a Deep Learning Library which supports Theano and TensorFlow at backend.
  • Keras is easier to install compared to other frameworks and had a better Documentation.
  • https://keras.io/

Data Model

Results and Outcome

  • A live video captures an image if a face detected in the frame.
  • The cropped image (face) is sent to the pipeline where the a prediction is made with already saved model which is loaded globally.
  • An additional layer (Sigmoid) is added at the flatten layer of the model.
  • The Prediction lies between 0-1.
  • Values near to 0 signifies Female. Values near 1 signifies Male.
  • A Cross Validation Accuracy of 90.92% is achieved on the Dataset trained.

Team Members

  • Mohit Reddy (mohitreddy1996)
  • Ambareesh Prakash (AeroDeath)
  • Vilas M Bhat (vilas897)
  • Aneesh Aithal (aneesh297)
  • Somnath Sarkar (somnathsarkar)

About

Gender Prediction using Deep Learning Algorithms and Model on Images of an Individual.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%