pytorch-deeplab-xception

TODO

• Basic deeplab v3+ model, using modified xception as backbone
• Training deeplab v3+ on Pascal VOC 2012, SBD, Cityscapes datasets
• Results evaluation on Pascal VOC 2012 test set
• Deeplab v3+ model using resnet as backbone

Introduction

This is a PyTorch(0.4.0) implementation of DeepLab-V3-Plus. It can use Modified Aligned Xception and ResNet as backbone. Currently, we train DeepLab V3 Plus using Pascal VOC 2012, SBD and Cityscapes datasets.

We trained deeplab v3+ (xception) on Pascal VOC 2012 and SBD datasets. After 50 training epoch, our deeplab v3+ model can reach 74.4% mIoU on Pascal VOC 2012 test set. More results will be available soon!

Installation

The ode was tested with Anaconda and Python 3.5. After installing the Anaconda environment:

0. Clone the repo:

   git clone https://github.com/jfzhang95/pytorch-deeplab-xception.git
   cd pytorch-deeplab-xception

1. Install dependencies:

   For PyTorch dependency, see pytorch.org for more details.

   For custom dependencies:

   pip install matplotlib pillow tensorboardX

2. Configure your dataset path in mypath.py.

3. You can train deeplab v3+ using xception or resnet as backbone.

   To train DeepLabV3+ on Pascal VOC 2012, please do:

   python train.py

   To train it on Cityscapes, please do:

   python train_cityscapes.py