Awesome Semantic Segmentation

Networks by architecture

• U-Net [https://arxiv.org/pdf/1505.04597.pdf]
  • https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/ [Caffe + Matlab]
  • https://github.com/jocicmarko/ultrasound-nerve-segmentation [Keras]
  • https://github.com/EdwardTyantov/ultrasound-nerve-segmentation [Keras]
  • https://github.com/ZFTurbo/ZF_UNET_224_Pretrained_Model [Keras]
  • https://github.com/yihui-he/u-net [Keras]
  • https://github.com/jakeret/tf_unet [Tensorflow]
• SegNet [https://arxiv.org/pdf/1511.00561.pdf]
  • https://github.com/alexgkendall/caffe-segnet [Caffe]
  • https://github.com/developmentseed/caffe/tree/segnet-multi-gpu [Caffe]
  • https://github.com/preddy5/segnet [Keras]
  • https://github.com/imlab-uiip/keras-segnet [Keras]
  • https://github.com/andreaazzini/segnet [Tensorflow]
  • https://github.com/fedor-chervinskii/segnet-torch [Torch]
  • https://github.com/0bserver07/Keras-SegNet-Basic [Keras]
  • https://github.com/tkuanlun350/Tensorflow-SegNet [Tensorflow]
• DeepLab [https://arxiv.org/pdf/1606.00915.pdf]
  • https://bitbucket.org/deeplab/deeplab-public/ [Caffe]
  • https://github.com/cdmh/deeplab-public [Caffe]
  • https://bitbucket.org/aquariusjay/deeplab-public-ver2 [Caffe]
  • https://github.com/TheLegendAli/DeepLab-Context [Caffe]
  • https://github.com/msracver/Deformable-ConvNets/tree/master/deeplab [MXNet]
  • https://github.com/DrSleep/tensorflow-deeplab-resnet [Tensorflow]
• Fully-Convolutional Network (FCN) [https://arxiv.org/pdf/1605.06211.pdf]
  • https://github.com/vlfeat/matconvnet-fcn [MatConvNet]
  • https://github.com/shelhamer/fcn.berkeleyvision.org [Caffe]
  • https://github.com/MarvinTeichmann/tensorflow-fcn [Tensorflow]
  • https://github.com/aurora95/Keras-FCN [Keras]
  • https://github.com/mzaradzki/neuralnets/tree/master/vgg_segmentation_keras [Keras]
  • https://github.com/k3nt0w/FCN_via_keras [Keras]
  • https://github.com/shekkizh/FCN.tensorflow [Tensorflow]
  • https://github.com/seewalker/tf-pixelwise [Tensorflow]
• ENet [https://arxiv.org/pdf/1606.02147.pdf]
  • https://github.com/TimoSaemann/ENet [Caffe]
  • https://github.com/e-lab/ENet-training [Torch]
  • https://github.com/PavlosMelissinos/enet-keras [Keras]
• LinkNet [https://arxiv.org/pdf/1707.03718.pdf]
  • https://github.com/e-lab/LinkNet [Torch]
• DenseNet [https://arxiv.org/pdf/1608.06993.pdf]
  • https://github.com/flyyufelix/DenseNet-Keras [Keras]
• Tiramisu [https://arxiv.org/pdf/1611.09326.pdf]
  • https://github.com/0bserver07/One-Hundred-Layers-Tiramisu [Keras]
• DilatedNet [https://arxiv.org/pdf/1511.07122.pdf]
  • https://github.com/nicolov/segmentation_keras [Keras]
• PixelNet [https://arxiv.org/pdf/1609.06694.pdf]
  • https://github.com/aayushbansal/PixelNet [Caffe]
• ICNet [https://arxiv.org/pdf/1704.08545.pdf]
  • https://github.com/hszhao/ICNet [Caffe]
• Mask-RCNN [https://arxiv.org/pdf/1703.06870.pdf]
  • https://github.com/CharlesShang/FastMaskRCNN [Tensorflow]
  • https://github.com/jasjeetIM/Mask-RCNN [Caffe]
• ERFNet [http://www.robesafe.uah.es/personal/eduardo.romera/pdfs/Romera17iv.pdf]
  • https://github.com/Eromera/erfnet [Torch]
• DeepMask [https://arxiv.org/pdf/1506.06204.pdf]
  • https://github.com/facebookresearch/deepmask [Torch]
• RefineNet [https://arxiv.org/pdf/1611.06612.pdf]
  • https://github.com/guosheng/refinenet [MatConvNet]
• PSPNet [https://arxiv.org/pdf/1612.01105.pdf]
  • https://github.com/hszhao/PSPNet [Caffe]
• CRFasRNN [http://www.robots.ox.ac.uk/%7Eszheng/papers/CRFasRNN.pdf]
  • https://github.com/torrvision/crfasrnn [Caffe]
  • https://github.com/sadeepj/crfasrnn_keras [Keras]
• Dilated convolution [https://arxiv.org/pdf/1511.07122.pdf]
  • https://github.com/fyu/dilation [Caffe]
• FCIS [https://arxiv.org/pdf/1611.07709.pdf]
  • https://github.com/msracver/FCIS [MxNet]

Networks by framework (Older list)

• Keras

  • https://github.com/gakarak/FCN_MSCOCO_Food_Segmentation
  • https://github.com/abbypa/NNProject_DeepMask

• TensorFlow

  • https://github.com/warmspringwinds/tf-image-segmentation

• Caffe

  • https://github.com/xiaolonw/nips14_loc_seg_testonly
  • https://github.com/naibaf7/caffe_neural_tool
  • http://cvlab.postech.ac.kr/research/deconvnet/

• torch

  • https://github.com/erogol/seg-torch
  • https://github.com/phillipi/pix2pix

• MXNet

  • https://github.com/tornadomeet/mxnet/tree/seg/example/fcn-xs
  • https://github.com/itijyou/ademxapp

Papers and Code (Older list)

• Simultaneous detection and segmentation

  • http://www.eecs.berkeley.edu/Research/Projects/CS/vision/shape/sds/
  • https://github.com/bharath272/sds_eccv2014

• Learning Deconvolution Network for Semantic Segmentation

  • https://github.com/HyeonwooNoh/DeconvNet

• Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation

  • https://github.com/HyeonwooNoh/DecoupledNet

• Learning to Propose Objects

  • http://vladlen.info/publications/learning-to-propose-objects/
  • https://github.com/philkr/lpo

• Nonparametric Scene Parsing via Label Transfer

  • http://people.csail.mit.edu/celiu/LabelTransfer/code.html

• Other

  • https://github.com/cvjena/cn24
  • http://lmb.informatik.uni-freiburg.de/resources/software.php
  • https://github.com/NVIDIA/DIGITS/tree/master/examples/semantic-segmentation
  • https://github.com/daijifeng001/MNC
  • https://github.com/voidrank/FastMask
  • http://jamie.shotton.org/work/code.html
  • https://github.com/amueller/textonboost

Graphical Models (CRF, MRF)

• https://github.com/cvlab-epfl/densecrf
• http://vladlen.info/publications/efficient-inference-in-fully-connected-crfs-with-gaussian-edge-potentials/
• http://www.philkr.net/home/densecrf
• http://graphics.stanford.edu/projects/densecrf/
• https://github.com/amiltonwong/segmentation/blob/master/segmentation.ipynb
• https://github.com/jliemansifry/super-simple-semantic-segmentation
• http://users.cecs.anu.edu.au/~jdomke/JGMT/
• https://www.quora.com/How-can-one-train-and-test-conditional-random-field-CRF-in-Python-on-our-own-training-testing-dataset
• https://github.com/tpeng/python-crfsuite
• https://github.com/chokkan/crfsuite
• https://sites.google.com/site/zeppethefake/semantic-segmentation-crf-baseline
• https://github.com/lucasb-eyer/pydensecrf

RNN

• https://github.com/fvisin/reseg
• https://github.com/bernard24/RIS
• https://github.com/martinkersner/train-CRF-RNN
• https://github.com/NP-coder/CLPS1520Project [Tensorflow]
• https://github.com/renmengye/rec-attend-public [Tensorflow]

Medical image segmentation:

• DIGITS

  • https://github.com/NVIDIA/DIGITS/tree/master/examples/medical-imaging

• U-Net: Convolutional Networks for Biomedical Image Segmentation

  • http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/
  • apache/mxnet#1514
  • https://github.com/orobix/retina-unet
  • https://github.com/fvisin/reseg
  • https://github.com/yulequan/melanoma-recognition
  • http://www.andrewjanowczyk.com/use-case-1-nuclei-segmentation/
  • https://github.com/junyanz/MCILBoost
  • https://github.com/imlab-uiip/lung-segmentation-2d
  • https://github.com/scottykwok/cervix-roi-segmentation-by-unet
  • https://github.com/WeidiXie/cell_counting_v2

• Cascaded-FCN

  • https://github.com/IBBM/Cascaded-FCN

• Keras

  • https://github.com/jocicmarko/ultrasound-nerve-segmentation
  • https://github.com/EdwardTyantov/ultrasound-nerve-segmentation
  • https://github.com/intact-project/ild-cnn

• Using Convolutional Neural Networks (CNN) for Semantic Segmentation of Breast Cancer Lesions (BRCA)

  • https://github.com/ecobost/cnn4brca

• Papers:

  • https://www2.warwick.ac.uk/fac/sci/dcs/people/research/csrkbb/tmi2016_ks.pdf
  • Sliding window approach
    • http://people.idsia.ch/~juergen/nips2012.pdf
  • https://github.com/albarqouni/Deep-Learning-for-Medical-Applications#segmentation

• Data:

  • https://luna16.grand-challenge.org/

Satellite images segmentation

• https://github.com/mshivaprakash/sat-seg-thesis
• https://github.com/KGPML/Hyperspectral
• https://github.com/lopuhin/kaggle-dstl
• https://github.com/mitmul/ssai
• https://github.com/mitmul/ssai-cnn
• https://github.com/azavea/raster-vision

• Data:
  • https://github.com/RSIA-LIESMARS-WHU/RSOD-Dataset-

Video segmentation

• https://github.com/shelhamer/clockwork-fcn

Autonomous driving

• https://github.com/MarvinTeichmann/MultiNet
• https://github.com/MarvinTeichmann/KittiSeg
• https://github.com/vxy10/p5_VehicleDetection_Unet [Keras]

Annotation Tools:

• https://github.com/AKSHAYUBHAT/ImageSegmentation
• https://github.com/kyamagu/js-segment-annotator
• https://github.com/CSAILVision/LabelMeAnnotationTool
• https://github.com/seanbell/opensurfaces-segmentation-ui
• https://github.com/lzx1413/labelImgPlus
• https://github.com/wkentaro/labelme

Loss functions

• Dice coefficient
• Jaccard loss
• sigmoid + binary crossentropy
• softmax + categorical crossentropy

Datasets:

• Stanford Background Dataset
• Sift Flow Dataset
• Barcelona Dataset
• Microsoft COCO dataset
• MSRC Dataset
• LITS Liver Tumor Segmentation Dataset
• KITTI
• Stanford background dataset
• Data from Games dataset
• Human parsing dataset
• Silenko person database
• Mapillary Vistas Dataset

Results:

• MSRC-21
• Cityscapes
• VOC2012

To look at

• keras-team/keras#6538
• https://github.com/warmspringwinds/tensorflow_notes
• https://github.com/meetshah1995/pytorch-semseg
• https://github.com/kjw0612/awesome-deep-vision#semantic-segmentation
• https://github.com/desimone/segmentation-models
• mrgloom#1
• https://github.com/nightrome/really-awesome-semantic-segmentation
• https://github.com/kjw0612/awesome-deep-vision#semantic-segmentation
• http://www.it-caesar.com/list-of-contemporary-semantic-segmentation-datasets/

Blog posts, other:

• https://handong1587.github.io/deep_learning/2015/10/09/segmentation.html
• http://www.andrewjanowczyk.com/efficient-pixel-wise-deep-learning-on-large-images/
• https://devblogs.nvidia.com/parallelforall/image-segmentation-using-digits-5/
• https://github.com/NVIDIA/DIGITS/tree/master/examples/binary-segmentation
• https://github.com/NVIDIA/DIGITS/tree/master/examples/semantic-segmentation