README.md

FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation

[Project] [Paper] [arXiv] [Home]

Official implementation of FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation.
A Faster, Stronger and Lighter framework for semantic segmentation, achieving the state-of-the-art performance and more than 3x acceleration.

@inproceedings{wu2019fastfcn,
  title     = {FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation},
  author    = {Wu, Huikai and Zhang, Junge and Huang, Kaiqi and Liang, Kongming and Yu Yizhou},
  booktitle = {arXiv preprint arXiv:1903.11816},
  year = {2019}
}

Contact: Hui-Kai Wu ([email protected])

Overview

Framework

Joint Pyramid Upsampling (JPU)

Install

1. PyTorch 1.0 (Note: The code is test in the environment with python=3.5, cuda=9.0)
2. Install FastFCN

   git clone https://github.com/wuhuikai/FastFCN.git
   cd FastFCN
   PATH=.:$PATH
   python setup.py install
   

3. Install Requirements

   nose
   tqdm
   scipy
   cython
   requests
   

Train and Test

PContext

python scripts/prepare_pcontext.py
cd experiments/segmentation

Method Backbone mIoU FPS Model Scripts EncNet ResNet-50 49.91 18.77

EncNet+JPU (ours) ResNet-50 51.05 37.56 GoogleDrive bash PSP ResNet-50 50.58 18.08

PSP+JPU (ours) ResNet-50 50.89 28.48 GoogleDrive bash DeepLabV3 ResNet-50 49.19 15.99

DeepLabV3+JPU (ours) ResNet-50 50.07 20.67 GoogleDrive bash EncNet ResNet-101 52.60 (MS) 10.51

EncNet+JPU (ours) ResNet-101 54.03 (MS) 32.02 GoogleDrive bash

ADE20K

python scripts/prepare_ade20k.py
cd experiments/segmentation

Training Set

Method Backbone mIoU (MS) Model Scripts EncNet ResNet-50 41.11

EncNet+JPU (ours) ResNet-50 42.75 GoogleDrive bash EncNet ResNet-101 44.65

EncNet+JPU (ours) ResNet-101 44.34 GoogleDrive bash

Training Set + Val Set

Method Backbone FinalScore (MS) Model Scripts EncNet+JPU (ours) ResNet-50

GoogleDrive bash EncNet ResNet-101 55.67

EncNet+JPU (ours) ResNet-101 55.84 GoogleDrive bash

Note: EncNet (ResNet-101) is trained with crop_size=576, while EncNet+JPU (ResNet-101) is trained with crop_size=480 for fitting 4 images into a 12G GPU.

Visual Results

Dataset Input GT EncNet Ours PContext ADE20K

More Visual Results

Acknowledgement

Code borrows heavily from PyTorch-Encoding.