This repo contains the code for a fierce attempt to implement this amazing Research paper using Keras.
In this repo you will find the all steps necessary to build a Semantic segmentation model. You will learn how to do the following:
- Reading images
- Image Preprocessing
- Data pipeline
- Building the Model
Any suggestions to improve this repository, including any new segmentation models you would like to see are welcome!
If you find this repository useful, please consider citing it using a link to the repo. :)
The getImages notebook has a step by step how to read images from any directory. For a more detailed guide you can check my article on medium BiSeNet for Real-Time Segmentation Part II.
The Preprocessing notebook has a full image preprocessing tutorial using opencv and includes getting segmentation masks. For a more detailed and personalized guide you can check my article on medium Image Pre-processing.
The Model is still under development. Based on the Bilateral Segmentation Network paper on arxiv: https://arxiv.org/abs/1808.00897
So far, only the BiSeNet model is done. Difficulties are:
- Auxiliary Loss function(softmax).