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 organized 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.py

The 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.

./job_local.sh YOUR_CS_PATH

Some recent projects have already benefited from ourimplementations. For example, Object Context Network(OCNet)[https://github.com/PkuRainBow/OCNet] currently achieves the state-of-the-art resultson Cityscapes and ADE20K. In addition, our code alsomake great contributions to Context Embedding with EdgePerceiving (CE2P), which won the 1st places in all hu-man parsing tracks in the 2nd LIP Challange.