Here is a tutorial on how to implement a research Paper with Keras
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:
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: