/Human-face-semantic-segmentation

A semantic segmentation for a human parsing task in Tensorflow Python

Primary LanguageJupyter Notebook

Human-face-semantic-segmentation

A semantic segmentation for a human parsing task in Tensorflow Python

This is the fourth-year project in the Electrical and Computer Engineering Department at Nazarbayev University.

This project uses the small portion of 4000 images from the Celeba dataset, which contains the faces of different celebrities. Source of the data: https://drive.google.com/drive/folders/1XvsWDwjvgg1No2zfzlTLMhL2JBi2qJSx?usp=sharing.

The DeepLabV3+ model was trained to classify the human face segmentation masks. You can read the report to better understand the problem and the results we've got.

You can access an already trained model in an .h5/.hdf5 format by clicking on the following link: https://drive.google.com/drive/folders/1qqKjS2KZpS4iKp8jeYoTJiNyg3E2BHrL?usp=sharing.

Good luck!