Pytorch code for semantic segmentation. This is a minimal code to run PSPnet and Deeplabv3 on Cityscape dataset. Shortly afterwards, the code will be reviewed and reorganized for convenience.
- Synchronous BN
- Fewness of Training Time
- Better Reproduced Performance
To install PyTorch>=0.4.0, please refer to https://github.com/pytorch/pytorch#installation.
Some parts of InPlace-ABN have a native CUDA implementation, which must be compiled with the following commands:
cd libs
sh build.sh
python build.pyThe build.sh script assumes that the nvcc compiler is available in the current system search path.
The CUDA kernels are compiled for sm_50, sm_52 and sm_61 by default.
To change this (e.g. if you are using a Kepler GPU), please edit the CUDA_GENCODE variable in build.sh.
Plesae download cityscapes dataset and unzip the dataset into YOUR_CS_PATH.
Please download MIT imagenet pretrained resnet101-imagenet.pth, and put it into dataset folder.
./run_local.sh YOUR_CS_PATHSome recent projects have already benefited from our implementations. For example, Object Context Network(OCNet) currently achieves the state-of-the-art resultson Cityscapes and ADE20K. In addition, our code also make great contributions to Context Embedding with EdgePerceiving (CE2P), which won the 1st places in all human parsing tracks in the 2nd LIP Challange.
If you find this code useful in your research, please consider citing:
@online{torch2018segment,
author = {Huang, Zilong and Wei, Yunchao and Wang, Xinggang, and Liu, Wenyu},
title = {A PyTorch Semantic Segmentation Toolbox},
year = 2018,
url = {https://github.com/speedinghzl/pytorch-segmentation-toolbox}
}