Skip to content
/ RBC Public

RBC: Rectifying the Biased Context in Continual Semantic Segmentation

Notifications You must be signed in to change notification settings

dcdcvgroup/RBC

Repository files navigation

RBC

RBC: Rectifying the Biased Context in Continual Semantic Segmentation ECCV2022.

Requirements

You need to install the following libraries:

  • Python (3.6)
  • Pytorch (1.8.1+cu102)
  • torchvision (0.9.1+cu102)
  • tensorboardX (1.8)
  • apex (0.1)
  • matplotlib (3.3.1)
  • numpy (1.17.2)
  • inplace-abn (1.0.7)

Note also that apex seems to only work with some CUDA versions, therefore try to install Pytorch (and torchvision) with the 10.2 CUDA version. You'll probably need anaconda instead of pip in that case:

conda install -y pytorch torchvision cudatoolkit=10.2 -c pytorch
cd apex
pip3 install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./

The default is to use a pretraining for the backbone used, that is searched in the pretrained folder of the project. We used the pretrained model released by the authors of In-place ABN (as said in the paper), that can be found here: link. You can also download the pretrained model by running the script download_resnet101_iabn_sync.sh from link.

How to perform training

We provide some scripts under the directory scripts/voc to reproduce the results in our paper (included tasks are 15-5, 15-1, 19-1 (VOC)).

For example, do

bash scripts/voc/rbc_15-1.sh

Note that you will need to modify those scripts to include the path where your data. And all of the provided scripts utilize the overlapped setting as default. If enable the disjoint setting, just delete the "--overlap".

Thanks

Our code is modified from PLOP.

Citation

@article{zhao2022rbc,
  title={RBC: Rectifying the Biased Context in Continual Semantic Segmentation},
  author={Zhao, Hanbin and Yang, Fengyu and Fu, Xinghe and Li, Xi},
  journal={arXiv preprint arXiv:2203.08404},
  year={2022}
}

About

RBC: Rectifying the Biased Context in Continual Semantic Segmentation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published