Awesome

⚠️ <b>WARNING:</b> This repository is responsible for the implementation of a multispectral image segmentation neural network used in <a href="https://github.com/acannistra/planet-snowcover">planet-snowcover</a>. We offer no warranty, express or implied, for functionality that suits any other purpose, including those indicated by the original repository and its owners. The following readme is kept for posterity, but much of the functionality is different. ⚠️

<h1 align='center'>RoboSat.pink</h1> <h2 align='center'>Semantic Segmentation ecosystem for GeoSpatial Imagery</h2> <p align=center> <img src="https://pbs.twimg.com/media/DpjonykWwAANpPr.jpg" alt="RoboSat pipeline extracting buildings from Imagery and Fusion" /> </p>

This repository is a DataPink flavor of RoboSat, including our latests developments.

Spirit:

Cutting edge Computer Vision research papers implementation
Industrial code robustness
Several tools, you can combine together (as Lego)
Extensible, by design
High performances
Minimalism as a code aesthetic
GeoSpatial standards compliancy
OSM and MapBox ecosystems friendly

Aims:

DataSet Quality Analysis
Change Detection highlighter
Features extraction and completion

Install:

1) Prerequisites:

Python >= 3.6 and PyTorch >= 0.4 installed, with related Nvidia GPU drivers, CUDA and CUDNN libs.
At least one GPU, with RAM GPU >= 6Go (default batch_size settings is targeted to 11Go).
Libs with headers: libjpeg, libwebp, libbz2, zlib, libboost. And Qt dependancies: libsm and libxrender. On a recent Ubuntu-server, could be done with:

    apt-get install build-essential libboost-python-dev zlib1g-dev libbz2-dev libjpeg-turbo8-dev libwebp-dev libsm6 libxrender1

2) Python libs Install:

     python3 -m pip install -r requirements.txt

NOTA: if you want to significantly increase performances switch from Pillow to <a href="https://github.com/uploadcare/pillow-simd">Pillow-simd</a>.

3) Deploy:

Move the rsp command to a bin directory covered by your PATH (or update your PATH)
Move the robosat_pink dir to somewhere covered by your PYTHONPATH (or update your PYTHONPATH)

WorkFlows:

Related resources:

<a href="./docs/from_opendata_to_opendataset.md">RoboSat.pink tutorial: from OpenData to OpenDataSet</a>
<a href="./docs/extensibility_by_design.md">RoboSat.pink documentation: Extensibility by Design</a>
<a href="http://www.datapink.com/presentations/2018-pyparis.pdf">Robosat slides @PyParis 2018</a>
<a href="https://github.com/mapbox/robosat">MapBox RoboSat github directory</a>
<a href="https://github.com/chrieke/awesome-satellite-imagery-datasets">Christoph Rieke's Awesome Satellite Imagery Datasets</a>
<a href="https://github.com/mrgloom/awesome-semantic-segmentation">Mr Gloom's Awesome Semantic Segmentation</a>

Bibliography:

<a href="http://www.cs.umanitoba.ca/~ywang/papers/isvc16.pdf">Optimizing IoU in Deep Neural Networks for Image Segmentation</a>
<a href="http://www.cs.toronto.edu/~wenjie/papers/iccv17/mattyus_etal_iccv17.pdf">DeepRoadMapper: Extracting Road Topology from Aerial Images</a>
<a href="https://arxiv.org/abs/1705.08790">The Lovász-Softmax loss: A tractable surrogate for the optimization of the IoU measure in neural networks</a>
<a href="https://arxiv.org/abs/1505.04597">U-Net: Convolutional Networks for Biomedical Image Segmentation</a>
<a href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>
<a href="https://arxiv.org/pdf/1804.08024.pdf">Angiodysplasia Detection and Localization Using Deep Convolutional Neural Networks</a>
<a href="https://arxiv.org/abs/1806.00844">TernausNetV2: Fully Convolutional Network for Instance Segmentation</a>
<a href="https://hal.archives-ouvertes.fr/hal-01523573/document">Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps</a>
<a href="https://arxiv.org/abs/1712.02616">In-Place Activated BatchNorm for Memory-Optimized Training of DNNs</a>

Arch:

Contributions and Services:

Pull Requests are welcome ! Feel free to send code... Don't hesitate either to initiate a prior discussion throught tickets on any implementation question.
If you want to collaborate through code production and maintenance on a long term basis, please get in touch, co-edition with an ad hoc governance can be considered.
If you want a new feature, but don't want to implement it, <a href="http://datapink.com">DataPink</a> provide core-dev services.
Expertise and training on RoboSat.pink are also provided by <a href="http://datapink.com">DataPink</a>.
And if you want to support the whole project, because it means for your own business, funding is also welcome.