We develop a new edge detection algorithm, holistically-nested edge detection (HED), which performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets. HED automatically learns rich hierarchical representations (guided by deep supervision on side responses) that are important in order to resolve the challenging ambiguity in edge and object boundary detection. We significantly advance the state-of-the-art on the BSD500 dataset (ODS F-score of .790) and the NYU Depth dataset (ODS F-score of .746), and do so with an improved speed (0.4s per image).

hed-master
hed-master\.Doxyfile
hed-master\.gitignore
hed-master\.travis.yml
hed-master\CMakeLists.txt
hed-master\LICENSE
hed-master\Makefile
hed-master\Makefile.config.example
hed-master\README.md
hed-master\caffe.cloc
hed-master\cmake
hed-master\cmake\ConfigGen.cmake
hed-master\cmake\Cuda.cmake
hed-master\cmake\Dependencies.cmake
hed-master\cmake\External
hed-master\cmake\External\gflags.cmake
hed-master\cmake\External\glog.cmake
hed-master\cmake\Misc.cmake
hed-master\cmake\Modules
hed-master\cmake\Modules\FindAtlas.cmake
hed-master\cmake\Modules\FindGFlags.cmake
hed-master\cmake\Modules\FindGlog.cmake
hed-master\cmake\Modules\FindLAPACK.cmake
hed-master\cmake\Modules\FindLMDB.cmake
hed-master\cmake\Modules\FindLevelDB.cmake
hed-master\cmake\Modules\FindMKL.cmake
hed-master\cmake\Modules\FindMatlabMex.cmake
hed-master\cmake\Modules\FindNumPy.cmake
hed-master\cmake\Modules\FindOpenBLAS.cmake
hed-master\cmake\Modules\FindSnappy.cmake
hed-master\cmake\Modules\FindvecLib.cmake
hed-master\cmake\ProtoBuf.cmake
hed-master\cmake\Summary.cmake
hed-master\cmake\Targets.cmake
hed-master\cmake\Templates
hed-master\cmake\Templates\CaffeConfig.cmake.in
hed-master\cmake\Templates\CaffeConfigVersion.cmake.in
hed-master\cmake\Templates\caffe_config.h.in
hed-master\cmake\Utils.cmake
hed-master\cmake\lint.cmake
hed-master\docs
hed-master\docs\CMakeLists.txt
hed-master\docs\CNAME
hed-master\docs\README.md
hed-master\docs\_config.yml
hed-master\docs\_layouts
hed-master\docs\_layouts\default.html
hed-master\docs\development.md
hed-master\docs\images
hed-master\docs\images\GitHub-Mark-64px.png
hed-master\docs\images\caffeine-icon.png
hed-master\docs\index.md
hed-master\docs\install_apt.md
hed-master\docs\install_osx.md
hed-master\docs\install_yum.md
hed-master\docs\installation.md
hed-master\docs\model_zoo.md
hed-master\docs\performance_hardware.md
hed-master\docs\stylesheets
hed-master\docs\stylesheets\pygment_trac.css
hed-master\docs\stylesheets\reset.css
hed-master\docs\stylesheets\styles.css
hed-master\docs\tutorial
hed-master\docs\tutorial\convolution.md
hed-master\docs\tutorial\data.md
hed-master\docs\tutorial\fig
hed-master\docs\tutorial\fig\.gitignore
hed-master\docs\tutorial\fig\backward.jpg
hed-master\docs\tutorial\fig\forward.jpg
hed-master\docs\tutorial\fig\forward_backward.png
hed-master\docs\tutorial\fig\layer.jpg
hed-master\docs\tutorial\fig\logreg.jpg
hed-master\docs\tutorial\forward_backward.md
hed-master\docs\tutorial\index.md
hed-master\docs\tutorial\interfaces.md
hed-master\docs\tutorial\layers.md
hed-master\docs\tutorial\loss.md
hed-master\docs\tutorial\net_layer_blob.md
hed-master\docs\tutorial\solver.md
hed-master\examples
hed-master\examples\hed
hed-master\examples\hed\HED-tutorial.ipynb
hed-master\examples\hed\deploy.prototxt
hed-master\examples\hed\solve.py
hed-master\examples\hed\solver.prototxt
hed-master\examples\hed\train_val.prototxt
hed-master\include
hed-master\include\caffe
hed-master\include\caffe\blob.hpp
hed-master\include\caffe\caffe.hpp
hed-master\include\caffe\common.hpp
hed-master\include\caffe\common_layers.hpp
hed-master\include\caffe\data_layers.hpp
hed-master\include\caffe\data_reader.hpp
hed-master\include\caffe\data_transformer.hpp
hed-master\include\caffe\filler.hpp
hed-master\include\caffe\internal_thread.hpp
hed-master\include\caffe\layer.hpp
hed-master\include\caffe\layer_factory.hpp
hed-master\include\caffe\loss_layers.hpp

• We are an exchange download platform that only provides communication channels. The downloaded content comes from the internet. Except for download issues, please Google on your own.
• The downloaded content is provided for members to upload. If it unintentionally infringes on your copyright, please contact us.
• Please use Winrar for decompression tools
• If download fail, Try it againg or Feedback to us.
• If downloaded content did not match the introduction, Feedback to us，Confirm and will be refund.
• Before downloading, you can inquire through the uploaded person information

Nothing．