0%

Segmentation Neural Network List Up

Migration to Meta Sheet


U-Net: Convolutional Networks for Biomedical Image Segmentation

MICCAI 2015


Paper

  • https://arxiv.org/pdf/1505.04597.pdf

Source Code Repository

  • Official Model: https://lmb.informatik.uni-freiburg.de/resources/opensource/unet/
    • Framework: Caffe
  • Reproduce Model: https://github.com/jakeret/tf_unet
    • Framework: Tensorflow
    • Paper: https://arxiv.org/pdf/1609.09077.pdf in Astronomy and Computing 2017
    • Documentation: https://tf-unet.readthedocs.io/en/latest/installation.html
  • Reproduce Model: https://github.com/zhixuhao/unet
    • Framework: Keras
  • Reproduce Model: https://github.com/milesial/Pytorch-UNet
    • Framework: PyTorch

Train Model Code, Test Model Code

  • Train Model Code
    • Reproduce Model(Tensorflow) Supports API
      • https://tf-unet.readthedocs.io/en/latest/usage.html
    • Reproduce Model(Keras)
      • https://github.com/zhixuhao/unet/blob/master/trainUnet.ipynb
  • Test Model Code:
    • Reproduce Model(Tensorflow) Supports API
      • https://tf-unet.readthedocs.io/en/latest/usage.html
    • Reproduce Model(Keras)
      • https://github.com/zhixuhao/unet/blob/master/trainUnet.ipynb

REMARK Reproduce Model(Keras) supports data augmentation docs

  • https://github.com/zhixuhao/unet/blob/master/dataPrepare.ipynb

Pre-Trained Model

  • Unsupported

Prediction Examples

UNET

Ref: https://arxiv.org/pdf/1505.04597.pdf

Citation

1
2
3
4
5
6
7
8
9
@article{akeret2017radio,
title={Radio frequency interference mitigation using deep convolutional neural networks},
author={Akeret, Joel and Chang, Chihway and Lucchi, Aurelien and Refregier, Alexandre},
journal={Astronomy and Computing},
volume={18},
pages={35--39},
year={2017},
publisher={Elsevier}
}



RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation

CVPR 2017


Paper

  • https://arxiv.org/pdf/1611.06612.pdf

Source Code Repository

  • Official Model: https://github.com/guosheng/refinenet
    • Framework: MATLAB
  • Reproduce Model: https://github.com/DrSleep/refinenet-pytorch
    • Framework: PyTorch
  • Reproduce Model: https://github.com/DrSleep/light-weight-refinenet
    • Light Weight RefineNet
    • Framework: PyTorch
  • Reproduce Model: https://github.com/eragonruan/refinenet-image-segmentation
    • Framework: Tensorflow

Train Model Code, Test Model Code

  • Train Model Code: Supported
  • Test Model Code: Supported

Pre-Trained Model

  • Supported
    • https://github.com/guosheng/refinenet/blob/master/libs/matconvnet/doc/site/docs/pretrained.md

Prediction Examples

REFINENET

Ref: https://arxiv.org/pdf/1611.06612.pdf

Ref: https://github.com/guosheng/refinenet

Citation

1
2
3
4
5
6
7
8
@inproceedings{Lin:2017:RefineNet,
title = {Refine{N}et: {M}ulti-Path Refinement Networks for High-Resolution Semantic Segmentation},
shorttitle = {RefineNet: Multi-Path Refinement Networks},
booktitle = {CVPR},
author = {Lin, G. and Milan, A. and Shen, C. and Reid, I.},
month = jul,
year = {2017}
}



PSPNet: Pyramid Scene Parsing Network

CVPR 2017, The Winner in 2016 ILSVRC Scene Parsing Challenge


Paper

  • https://arxiv.org/pdf/1612.01105.pdf

Source Code Repository

  • Official Model: https://github.com/hszhao/PSPNet
    • Framework: Caffe
  • Reproduce Model: https://github.com/Vladkryvoruchko/PSPNet-Keras-tensorflow
    • Framework: Keras
  • Reproduce Model: https://github.com/hellochick/PSPNet-tensorflow
    • Framework: Tensorflow

Train Model Code, Test Model Code

  • Train Model Code: Supported in Reproduce Model
    • https://github.com/Vladkryvoruchko/PSPNet-Keras-tensorflow
  • Test Model Code: Supported in Reproduce Model
    • https://github.com/Vladkryvoruchko/PSPNet-Keras-tensorflow

Pre-Trained Model

  • Supported
    • Offical Model
      • https://github.com/hszhao/PSPNet
  • Supported
    • Reproduce Model
      • https://github.com/Vladkryvoruchko/PSPNet-Keras-tensorflow

Prediction Examples

Ref: https://hszhao.github.io/projects/pspnet/

Citation

1
2
3
4
5
6
7
8
9
10
@inproceedings{zhao2017pspnet,
author = {Hengshuang Zhao and
Jianping Shi and
Xiaojuan Qi and
Xiaogang Wang and
Jiaya Jia},
title = {Pyramid Scene Parsing Network},
booktitle = {Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2017}
}



Large Kernel Matters: Improve Semantic Segmentation by Global Convolutional Network

CVPR 2017


Paper

  • https://arxiv.org/pdf/1612.01105.pdf

Source Code Repository

  • Official Model: None

REMARK Large Kernel Matters를 Tensorflow로 구현한 코드 및 설명이 기술된 한국 블로그

  • http://research.sualab.com/practice/2018/11/23/image-segmentation-deep-learning.html

Train Model Code, Test Model Code

  • Train Model Code: Unsupported
  • Test Model Code: Unsupported

Pre-Trained Model

  • Unsupported

Prediction Examples

GCN

Ref: https://arxiv.org/pdf/1703.02719.pdf




DeepLab v3(+): Atrous SeparableConvolution for Semantic Image Segmentation

ECCV 2018


Paper

  • DeepLab v3: https://arxiv.org/pdf/1706.05587.pdf
  • DeepLab v3+: https://arxiv.org/pdf/1802.02611.pdf

Source Code Repository

  • Official Model: https://github.com/tensorflow/models/tree/master/research/deeplab
    • Framework: Tensorflow

Train Model Code, Test Model Code

  • Train Model Code: Supported
  • Test Model Code: Supported

Pre-Trained Model

  • Supported

Prediction Examples

Ref: https://github.com/tensorflow/models/tree/master/research/deeplab

Citation

1
2
3
4
5
6
@inproceedings{deeplabv3plus2018,
title={Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation},
author={Liang-Chieh Chen and Yukun Zhu and George Papandreou and Florian Schroff and Hartwig Adam},
booktitle={ECCV},
year={2018}
}



Inplace ABN: In-Place Activated BatchNorm for Memory-Optimized Training of DNNs

CVPR 2018


Paper

  • https://arxiv.org/pdf/1712.02616v3.pdf

Source Code Repository

  • Official Model: https://github.com/mapillary/inplace_abn
    • Framework: PyTorch

Train Model Code, Test Model Code

  • Train Model Code: Supported
  • Test Model Code: Supported

Pre-Trained Model

  • Supported

Citation

1
2
3
4
5
6
@inproceedings{rotabulo2017place,
title={In-Place Activated BatchNorm for Memory-Optimized Training of DNNs},
author={Rota Bul\`o, Samuel and Porzi, Lorenzo and Kontschieder, Peter},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
year={2018}
}



TernausNetV2: Fully Convolutional Network for Instance Segmentation

2nd place in CVPR 2018 DeepGlobe Building Extraction Challenge


Paper

  • http://openaccess.thecvf.com/content_cvpr_2018_workshops/papers/w4/Iglovikov_TernausNetV2_Fully_Convolutional_CVPR_2018_paper.pdf

Source Code Repository

  • Official Model: https://github.com/ternaus/TernausNetV2
    • Framework: PyTorch

Train Model Code, Test Model Code

  • Train Model Code: Unsupported
  • Test Model Code: Unsupported

Pre-Trained Model

  • Unsupported

Prediction Example

Ref: https://github.com/ternaus/TernausNetV2

Citation

1
2
3
4
5
6
7
@InProceedings{Iglovikov_2018_CVPR_Workshops,
author = {Iglovikov, Vladimir and Seferbekov, Selim and Buslaev, Alexander and Shvets, Alexey},
title = {TernausNetV2: Fully Convolutional Network for Instance Segmentation},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2018}
}



Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning

ICMLA 2018, Wining solution and its improvement for MICCAI 2017 Robotic Instrument Segmentation Sub-Challenge


Paper

  • https://arxiv.org/pdf/1803.01207.pdf

Source Code Repository

  • Official Model: https://github.com/ternaus/TernausNetV2
    • Framework: PyTorch

Train Model Code, Test Model Code

  • Train Model Code: Supported
  • Test Model Code: Supported

Pre-Trained Model

  • Supported in https://drive.google.com/drive/folders/13e0C4fAtJemjewYqxPtQHO6Xggk7lsKe

Prediction Example

Ref: https://github.com/ternaus/robot-surgery-segmentation

Citation

1
2
3
4
5
6
@article{shvets2018automatic,
title={Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning},
author={Shvets, Alexey and Rakhlin, Alexander and Kalinin, Alexandr A and Iglovikov, Vladimir},
journal={arXiv preprint arXiv:1803.01207},
year={2018}
}


//