# PSPNet-Pytorch

**Repository Path**: zhangcaocao/PSPNet-Pytorch

## Basic Information

- **Project Name**: PSPNet-Pytorch
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-03-05
- **Last Updated**: 2021-06-21

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# PSPNet-Pytorch
This Repo contains an implemetation of "Pyramid Scene Parsing Network" in Pytorch. Pretrained weights are converted from Official Caffe Repo. The performance is:
[acc 0.95196], [acc_cls 0.83369], [mean_iu 0.75547], [fwavacc 0.91172]
on Validation set of CityScapes Dataset.

## Installation
1. Install Anaconda3 from [Here](https://www.anaconda.com)
2. Create a Conda Environment: 
```console
username@PC:~$ conda env create -f environment.yml
```
3. Download Cityscapes Dataset from [Here](https://www.cityscapes-dataset.com/)
4. Download Caffe Pretrained from [Here](https://drive.google.com/open?id=0BzaU285cX7TCT1M3TmNfNjlUeEU) and put it in Caffe-PSPNet folder
4. Run the code:
```console
username@PC:~$ python main.py
```


## Qualitative 
<p float="left">
  <img src="prediction/1.png" width="250" />
  <img src="prediction/2.png" width="250" /> 
  <img src="prediction/3.png" width="250" />
</p>


## Note
Weight conversion from Caffe to Pytorch is modified from [pytorch-semseg](https://github.com/meetshah1995/pytorch-semseg). <br />
Preprocessing and loss function is modified from [pytorch-semantic-segmentation](https://github.com/zijundeng/pytorch-semantic-segmentation). <br />
Since the differences are significant, I decided not to add a branch to any of the the above project.