.. _sphx_glr_build_examples_segmentation_demo_fcn.py:

1. Getting Started with FCN Pre-trained Models
==============================================

This is a quick demo of using GluonCV FCN model.
Please follow the `installation guide <../index.html>`_ to install MXNet and GluonCV if not yet.



.. code-block:: python

    import mxnet as mx
    from mxnet import image
    from mxnet.gluon.data.vision import transforms
    import gluoncv
    # using cpu
    ctx = mx.cpu(0)








Prepare the image
-----------------

download the example image



.. code-block:: python

    url = 'https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/1.jpg'
    filename = 'example.jpg'
    gluoncv.utils.download(url, filename)







load the image



.. code-block:: python

    with open(filename, 'rb') as f:
        img = image.imdecode(f.read())

    from matplotlib import pyplot as plt
    plt.imshow(img.asnumpy())
    plt.show()




.. image:: /build/examples_segmentation/images/sphx_glr_demo_fcn_001.png
    :align: center




normalize the image using dataset mean



.. code-block:: python

    transform_fn = transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize([.485, .456, .406], [.229, .224, .225])
    ])
    img = transform_fn(img)
    img = img.expand_dims(0).as_in_context(ctx)







Load the pre-trained model and make prediction
----------------------------------------------

get pre-trained model



.. code-block:: python

    model = gluoncv.model_zoo.get_model('fcn_resnet101_voc', pretrained=True)







make prediction using single scale



.. code-block:: python

    output = model.evaluate(img)
    predict = mx.nd.squeeze(mx.nd.argmax(output, 1)).asnumpy()







Add color pallete for visualization



.. code-block:: python

    from gluoncv.utils.viz import get_color_pallete
    import matplotlib.image as mpimg
    mask = get_color_pallete(predict, 'pascal_voc')
    mask.save('output.png')







show the predicted mask



.. code-block:: python

    mmask = mpimg.imread('output.png')
    plt.imshow(mmask)
    plt.show()




.. image:: /build/examples_segmentation/images/sphx_glr_demo_fcn_002.png
    :align: center




More Examples
 -------------

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/4.jpg
    :width: 45%

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/4.png
    :width: 45%

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/5.jpg
    :width: 45%

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/5.png
    :width: 45%

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/6.jpg
    :width: 45%

.. image:: https://raw.githubusercontent.com/zhanghang1989/gluon-vision-figures/master/voc_examples/6.png
    :width: 45%


**Total running time of the script:** ( 0 minutes  5.013 seconds)



.. only :: html

 .. container:: sphx-glr-footer


  .. container:: sphx-glr-download

     :download:`Download Python source code: demo_fcn.py <demo_fcn.py>`



  .. container:: sphx-glr-download

     :download:`Download Jupyter notebook: demo_fcn.ipynb <demo_fcn.ipynb>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.readthedocs.io>`_