Skip to content

Semantic segmentation in keras. Models used: FCN8, FCN32, UNET, SegNet. Balanced and unbalanced loss functions.

License

Notifications You must be signed in to change notification settings

nayemabs/keras_segmentation

Repository files navigation

This project is no longer updated. There is a confusing place, please refer to issues 5 and so on.Thank you for your support.
中文说明

Environment

win10;python3.6.7 ;

Item Value
keras 2.2.4
tensorflow-gpu 1.10.0

Reference:
https://github.com/divamgupta/image-segmentation-keras
https://github.com/ykamikawa/SegNet

Data:
https://drive.google.com/file/d/0B0d9ZiqAgFkiOHR1NTJhWVJMNEU/view?usp=sharing
Data or:
https://github.com/alexgkendall/SegNet-Tutorial/tree/master/CamVid

Project Strcutre

在这里插入图片描述

python visualizeDataset.py: Visual samples
在这里插入图片描述

python train.py: Execution train
python predict.py: Execution predict

You can modify the parameter in project switching model or cloning the historical version.

About

FCN32

Visualization results:
在这里插入图片描述

FCN8

Visualization results:
在这里插入图片描述

SegNet

U-Net

Not exactly in accordance with the realization of the paper.This project mainly helps to learn semantics segmentation model,and some data processing skills.

About

Semantic segmentation in keras. Models used: FCN8, FCN32, UNET, SegNet. Balanced and unbalanced loss functions.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages