Data

Tools

Indexes

Applications

Experiments

Open Source Science

samseg

SAM-Guided Domain Adaptation for Remote Sensing Segmentation

https://github.com/mengyao-wu/samseg

Science Score: 67.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ✓
    DOI references
    Found 2 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (7.2%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

SAM-Guided Domain Adaptation for Remote Sensing Segmentation

Basic Info
  • Host: GitHub
  • Owner: Mengyao-Wu
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Size: 24.4 KB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created 7 months ago · Last pushed 7 months ago
Metadata Files
Readme License Citation

README.md

SAMseg

This repo is the implementation of "Self-Training Guided Disentangled Adaptation for Cross-Domain Remote Sensing Image Semantic Segmentation". we refer to MMSegmentation and MMGeneration. Many thanks to SenseTime and their two excellent repos.

Dataset Preparation

We select Postsdam, Vaihingen and LoveDA as benchmark datasets and create train, val, test list for researchers to follow.

In the following, we provide the detailed commands for dataset preparation.

Potsdam

 Move the ‘3_Ortho_IRRG.zip’ and ‘5_Labels_all_noBoundary.zip’ to Potsdam_IRRG folder 
 Move the ‘2_Ortho_RGB.zip’ and ‘5_Labels_all_noBoundary.zip’ to Potsdam_RGB folder
 python tools/convert_datasets/potsdam.py yourpath/ST-DASegNet/data/Potsdam_IRRG/ --clip_size 512 --stride_size 512
 python tools/convert_datasets/potsdam.py yourpath/ST-DASegNet/data/Potsdam_RGB/ --clip_size 512 --stride_size 512

SeaElm:
python tools/convert_datasets/potsdam.py ./data/Potsdam_IRRG_DA/ -o=./data/Potsdam_IRRG_DA/ --clip_size=512 --stride_size=512
python tools/convert_datasets/potsdam.py ./data/Potsdam_RGB_DA/ -o=./data/Potsdam_RGB_DA/ --clip_size=512 --stride_size=512

Vaihingen

 Move the 'ISPRS_semantic_labeling_Vaihingen.zip' and 'ISPRS_semantic_labeling_Vaihingen_ground_truth_eroded_COMPLETE.zip' to Vaihingen_IRRG folder 
 python tools/convert_datasets/vaihingen.py yourpath/ST-DASegNet/data/Vaihingen_IRRG/ --clip_size 512 --stride_size 256

SeaElm:
python tools/convert_datasets/vaihingen.py /data2/yrz/data/Vaihingen/IRRG/ -o=/data2/yrz/data/Vaihingen/IRRG --clip_size 512 --stride_size 256

LoveDA

 Unzip Train.zip, Val.zip, Test.zip and create Train, Val and Test list for Urban and Rural

SAMSeg

Install

  1. requirements:

    python >= 3.7

    pytorch >= 1.4

    cuda >= 10.0

  2. prerequisites: Please refer to MMSegmentation PREREQUISITES.

    ``` cd SAMSeg

    pip install -e .

    chmod 777 ./tools/dist_train.sh

    chmod 777 ./tools/dist_test.sh ```

Training

mit_b5.pth : url For SegFormerb5 based ST-DASegNet training, we provide ImageNet-pretrained backbone here.

We select deeplabv3 and Segformerb5 as baselines. Actually, we use deeplabv3+, which is a more advanced version of deeplabv3. After evaluating, we find that deeplabv3+ has little modification compared to deeplabv3 and has little advantage than deeplabv3.

For LoveDA results, we evaluate on test datasets and submit to online server (https://github.com/Junjue-Wang/LoveDA) (https://codalab.lisn.upsaclay.fr/competitions/424). We also provide the evaluation results on validation dataset.

LoveDA UDA Urban LoveDA UDA Rural
  1. Potsdam IRRG to Vaihingen IRRG:

train code ```shell cd ST-DASegNet conda ./tools/disttrain.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940k_Potsdam2Vaihingen.py 2

./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixlow/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py 2 295073 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixcocomask/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py 1 295073 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py 2 295083 # work 66.49

--resume-from /data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Potsdam2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomask/iter_22500.pth

第三个点

./tools/disttrain.sh ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240k_Potsdam2Vaihingen.py 2 295103

CUDAVISIBLEDEVICES=0 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py \ --resume-from '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Potsdam2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kBase/iter_2500.pth'

CUDAVISIBLEDEVICES=0 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py \ --resume-from '' \

CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ --resume-from '' \

CUDAVISIBLEDEVICES=3 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ --resume-from '' \

""" **test code** shell

test

patchmixcocomask/iter_2500 ： 49.31

python tools/test.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Potsdam2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomask/iter2500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

patchmixcocomask/iter_22500 ：64.24

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomask/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py \ /data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Potsdam2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomask/iter22500.pth \ --eval mIoU mFscore --gpu-id 1 --auto-show

full

python tools/test.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kPotsdam2Vaihingen.py \ /data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Potsdam2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kfull/iter27500.pth \ --eval mIoU mFscore --gpu-id 2 --auto-show

  1. Vaihingen IRRG to Potsdam IRRG:

train code ```shell cd SAMSeg

./tools/disttrain.sh ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 2 294983 ./tools/disttrain.sh ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 2 294993 ./tools/disttrain.sh ./experiments/deeplabv3plus/cowmix/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 2 294003 ./tools/disttrain.sh ./experiments/deeplabv3plus/cowout/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 1 294013 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmix/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 1 294023 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchout/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 3 294033 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixdis/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 1 294043 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixtea/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 3 294053

./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixcocomask/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 3 294063 # work, 57.89 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixcocomaskdis/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 1 294073 ./tools/disttrain.sh ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py 1 294083

./tools/disttest.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940kPotsdam2Vaihingen.py 2 ./experiments/segformerb5/ST-DASegNetresults/P2VIRRG64.33.pth --eval mIoU
**test code** shell

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomask/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2Potsdamresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomask/iter2500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kVaihingen2Potsdam.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2Potsdamresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240random/iter_50000.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

```

  1. Potsdam RGB to Vaihingen IRRG: train ```shell cd SAMSeg

./tools/disttrain.sh ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py 0 294983 CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/patchmix/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/cowmix/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240k_PotsdamRGB2Vaihingen.py

CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240k_PotsdamRGB2Vaihingen.py **test** shell

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter2500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

python tools/test.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kBase/iter15000.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kfull/iter22500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter12500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240normalize2regionrelateentropy&impurityk=1/iter_20000.pth' \ --eval mIoU mFscore --gpu-id 3 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kPotsdamRGB2Vaihingen.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/PotsdamRGB2Vaihingenresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240normalize2regionrelateentropy&impurityk=1/iter_22500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

```

  1. Vaihingen IRRG to Potsdam RGB: train ```shell cd SAMSeg

./tools/disttrain.sh ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py 1 294993 CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py CUDAVISIBLEDEVICES=3 python tools/train.py ./experiments/deeplabv3plus/patchmix/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py CUDAVISIBLEDEVICES=3 python tools/train.py ./experiments/deeplabv3plus/cowmix/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py CUDAVISIBLEDEVICES=3 python tools/train.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240k_Vaihingen2PotsdamRGB.py

CUDAVISIBLEDEVICES=3 python tools/train.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240k_Vaihingen2PotsdamRGB.py ```

test ```shell

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data2/wmy/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter30000.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

python tools/test.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresults/deeplabv3plusr50-d84x4512x51240kBase/iter15000.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresults/deeplabv3plusr50-d84x4512x51240kfull/iter22500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter15000.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/cowmix/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresults/deeplabv3plusr50-d84x4512x51240kcowmix/iter12500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240regionrelateentropy&impurity/iter_40000.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kVaihingen2PotsdamRGB.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/Vaihingen2PotsdamRGBresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240regionrelateentropy&impurity/iter_22500.pth' \ --eval mIoU mFscore --gpu-id 0 --auto-show

```

  1. LoveDA Rural to Urban

```shell cd SAMSeg

./tools/disttrain.sh ./experiments/deeplabv3/configLoveDA/ST-DASegNetdeeplabv3plusr50-d84x4512x51240kR2U.py 2 ./tools/disttrain.sh ./experiments/segformerb5/configLoveDA/ST-DASegNetsegformerb5769x76940kR2U.py 2

CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4512x51240kR2U.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/patchmix/deeplabv3plusr50-d84x4512x51240kR2U.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kR2U.py CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4512x51240k_R2U.py

CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/segformerb5/configLoveDA/full/segformerb5769x76940kR2U.py

CUDAVISIBLEDEVICES=2 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4512x51240kR2U.py \ --resume-from '' \

CUDAVISIBLEDEVICES=0 python tools/train.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4768x76840kR2U.py \ --resume-from '' \

python tools/test.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4768x76840kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresults/deeplabv3plusr50-d84x4512x51240kBase/iter_7500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

python tools/test.py ./experiments/deeplabv3plus/base/deeplabv3plusr50-d84x4768x76840kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresults/deeplabv3plusr50-d84x4512x51240kBase/iter_37500.pth' \ --eval mIoU mFscore --gpu-id 4 --auto-show

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4768x76840kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter_37500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

python tools/test.py ./experiments/deeplabv3plus/patchmixcocomasktea/deeplabv3plusr50-d84x4768x76840kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresults/deeplabv3plusr50-d84x4512x51240kpatchmixcocomasktea/iter_7500.pth' \ --eval mIoU mFscore --gpu-id 3 --auto-show

python tools/test.py ./experiments/deeplabv3plus/adabase/deeplabv3plusr50-d84x4768x76840kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresultsada/deeplabv3plusr50-d84x4512x51240kADA0.02240normalize2regionrelateentropy&impurityk=1/iter37500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

python tools/test.py ./experiments/deeplabv3plus/full/deeplabv3plusr50-d84x4512x51240kR2U.py \ '/data/yrz/repos/ST-DASegNet/checkpoints/deeplabv3plus/LoveDAR2Uresults/deeplabv3plusr50-d84x4512x51240kfull/iter_37500.pth' \ --eval mIoU mFscore --gpu-id 2 --auto-show

```

Testing

Trained with the above commands, you can get a trained model to test the performance of your model.

  1. Testing commands

    ``` cd SAMSeg

    ./tools/disttest.sh yourpath/config.py yourpath/trainedmodel.pth --eval mIoU
    ./tools/dist    test.sh yourpath/config.py yourpath/trainedmodel.pth --eval mFscore ```

  2. Testing cases: P2VIRRG64.33.pth and V2PIRRG59.65.pth : google drive

    ``` cd SAMSeg

    ./tools/disttest.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940kPotsdam2Vaihingen.py 2 ./experiments/segformerb5/ST-DASegNetresults/P2VIRRG64.33.pth --eval mIoU
    ./tools/disttest.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940kPotsdam2Vaihingen.py 2 ./experiments/segformerb5/ST-DASegNetresults/P2VIRRG64.33.pth --eval mFscore ```

    ``` cd SAMSeg

    ./tools/disttest.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940kVaihingen2Potsdam.py 2 ./experiments/segformerb5/ST-DASegNetresults/V2PIRRG59.65.pth --eval mIoU
    ./tools/disttest.sh ./experiments/segformerb5/config/ST-DASegNetsegformerb5769x76940kVaihingen2Potsdam.py 2 ./experiments/segformerb5/ST-DASegNetresults/V2PIRRG59.65.pth --eval mFscore ```

The ArXiv version of this paper is release. ST-DASegNet_arxiv. This paper has been published on JAG, please refer to Self-Training Guided Disentangled Adaptation for Cross-Domain Remote Sensing Image Semantic Segmentation.

If you have any question, please discuss with me by sending email to 24171214037@xidian.edu.cn.

References

Many thanks to their excellent works * MMSegmentation * MMGeneration * DAFormer

Owner

  • Login: Mengyao-Wu
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
  - name: "MMSegmentation Contributors"
title: "OpenMMLab Semantic Segmentation Toolbox and Benchmark"
date-released: 2020-07-10
url: "https://github.com/open-mmlab/mmsegmentation"
license: Apache-2.0

GitHub Events

Total
  • Push event: 8
  • Create event: 1
Last Year
  • Push event: 8
  • Create event: 1

Dependencies

mmsegmentation.egg-info/requires.txt pypi
  • cityscapesscripts *
  • codecov *
  • flake8 *
  • interrogate *
  • isort ==4.3.21
  • matplotlib *
  • numpy *
  • packaging *
  • prettytable *
  • pytest *
  • xdoctest >=0.10.0
  • yapf *
requirements.txt pypi
setup.py pypi