/keras-fcn

A playable implementation of Fully Convolutional Networks with Keras.

Primary LanguagePythonMIT LicenseMIT

keras-fcn

Build Status codecov License: MIT

A re-implementation of Fully Convolutional Networks with Keras

Installation

Dependencies

  1. keras
  2. tensorflow

Install with pip

$ pip install git+https://github.com/JihongJu/keras-fcn.git

Build from source

$ git clone https://github.com/JihongJu/keras-fcn.git
$ cd keras-fcn
$ pip install --editable .

Usage

FCN with VGG16

from keras_fcn import FCN
fcn_vgg16 = FCN(input_shape=(500, 500, 3), classes=21,  
                weights='imagenet', trainable_encoder=True)
fcn_vgg16.compile(optimizer='rmsprop',
                  loss='categorical_crossentropy',
                  metrics=['accuracy'])
fcn_vgg16.fit(X_train, y_train, batch_size=1)

FCN with VGG19

from keras_fcn import FCN
fcn_vgg19 = FCN_VGG19(input_shape=(500, 500, 3), classes=21,  
                      weights='imagenet', trainable_encoder=True)
fcn_vgg19.compile(optimizer='rmsprop',
                  loss='categorical_crossentropy',
                  metrics=['accuracy'])
fcn_vgg19.fit(X_train, y_train, batch_size=1)

Custom FCN (VGG16 as an example)

from keras.layers import Input
from keras.models import Model
from keras_fcn.encoders import Encoder
from keras_fcn.decoders import VGGUpsampler
from keras_fcn.blocks import (vgg_conv, vgg_fc)
inputs = Input(shape=(224, 224, 3))
blocks = [vgg_conv(64, 2, 'block1'),
          vgg_conv(128, 2, 'block2'),
          vgg_conv(256, 3, 'block3'),
          vgg_conv(512, 3, 'block4'),
          vgg_conv(512, 3, 'block5'),
          vgg_fc(4096)]
encoder = Encoder(inputs, blocks, weights='imagenet',
                  trainable=True)
feat_pyramid = encoder.outputs   # A feature pyramid with 5 scales
feat_pyramid = feat_pyramid[:3]  # Select only the top three scale of the pyramid
feat_pyramid.append(inputs)      # Add image to the bottom of the pyramid


outputs = VGGUpsampler(feat_pyramid, scales=[1, 1e-2, 1e-4], classes=21)
outputs = Activation('softmax')(outputs)

fcn_custom = Model(inputs=inputs, outputs=outputs)

And implement a custom Fully Convolutional Network becomes simply define a series of convolutional blocks that one stacks on top of another.

Custom decoders

from keras_fcn.blocks import vgg_upsampling
from keras_fcn.decoders import Decoder
decode_blocks = [
vgg_upsampling(classes=21, target_shape=(None, 14, 14, None), scale=1),            
vgg_upsampling(classes=21, target_shape=(None, 28, 28, None),  scale=0.01),
vgg_upsampling(classes=21, target_shape=(None, 224, 224, None),  scale=0.0001)
]
outputs = Decoder(feat_pyramid[-1], decode_blocks)

The decode_blocks can be customized as well.

from keras_fcn.layers import BilinearUpSampling2D

def vgg_upsampling(classes, target_shape=None, scale=1, block_name='featx'):
    """A VGG convolutional block with bilinear upsampling for decoding.

    :param classes: Integer, number of classes
    :param scale: Float, scale factor to the input feature, varing from 0 to 1
    :param target_shape: 4D Tuples with targe_height, target_width as
    the 2nd, 3rd elements if `channels_last` or as the 3rd, 4th elements if
    `channels_first`.

    >>> from keras_fcn.blocks import vgg_upsampling
    >>> feat1, feat2, feat3 = feat_pyramid[:3]
    >>> y = vgg_upsampling(classes=21, target_shape=(None, 14, 14, None),
    >>>                    scale=1, block_name='feat1')(feat1, None)
    >>> y = vgg_upsampling(classes=21, target_shape=(None, 28, 28, None),
    >>>                    scale=1e-2, block_name='feat2')(feat2, y)
    >>> y = vgg_upsampling(classes=21, target_shape=(None, 224, 224, None),
    >>>                    scale=1e-4, block_name='feat3')(feat3, y)

    """
    def f(x, y):
        score = Conv2D(filters=classes, kernel_size=(1, 1),
                       activation='linear',
                       padding='valid',
                       kernel_initializer='he_normal',
                       name='score_{}'.format(block_name))(x)
        if y is not None:
            def scaling(xx, ss=1):
                return xx * ss
            scaled = Lambda(scaling, arguments={'ss': scale},
                            name='scale_{}'.format(block_name))(score)
            score = add([y, scaled])
        upscore = BilinearUpSampling2D(
            target_shape=target_shape,
            name='upscore_{}'.format(block_name))(score)
        return upscore
    return f

Try Examples

  1. Download VOC2011 dataset
$ wget "http://host.robots.ox.ac.uk/pascal/VOC/voc2011/VOCtrainval_25-May-2011.tar"
$ tar -xvzf VOCtrainval_25-May-2011.tar
$ mkdir ~/Datasets
$ mv TrainVal/VOCdevkit/VOC2011 ~/Datasets
  1. Mount dataset from host to container and start bash in container image

From repository keras-fcn

$ nvidia-docker run -it --rm -v `pwd`:/root/workspace -v ${Home}/Datasets/:/root/workspace/data jihong/keras-gpu bash

or equivalently,

$ make bash
  1. Within the container, run the following codes.
$ cd ~/workspace
$ pip setup.py -e .
$ cd voc2011
$ python train.py

More details see source code of the example in Training Pascal VOC2011 Segmention

Model Architecture

FCN8s with VGG16 as base net:

fcn_vgg16

TODO

  • Add ResNet