official/vision/modeling/backbones/resnet_deeplab.py

# Copyright 2023 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""Contains definitions of Residual Networks with Deeplab modifications."""

from typing import Callable, Optional, Tuple, List

import numpy as np
import tensorflow as tf, tf_keras
from official.modeling import hyperparams
from official.modeling import tf_utils
from official.vision.modeling.backbones import factory
from official.vision.modeling.layers import nn_blocks
from official.vision.modeling.layers import nn_layers

layers = tf_keras.layers

# Specifications for different ResNet variants.
# Each entry specifies block configurations of the particular ResNet variant.
# Each element in the block configuration is in the following format:
# (block_fn, num_filters, block_repeats)
RESNET_SPECS = {
    50: [
        ('bottleneck', 64, 3),
        ('bottleneck', 128, 4),
        ('bottleneck', 256, 6),
        ('bottleneck', 512, 3),
    ],
    101: [
        ('bottleneck', 64, 3),
        ('bottleneck', 128, 4),
        ('bottleneck', 256, 23),
        ('bottleneck', 512, 3),
    ],
    152: [
        ('bottleneck', 64, 3),
        ('bottleneck', 128, 8),
        ('bottleneck', 256, 36),
        ('bottleneck', 512, 3),
    ],
    200: [
        ('bottleneck', 64, 3),
        ('bottleneck', 128, 24),
        ('bottleneck', 256, 36),
        ('bottleneck', 512, 3),
    ],
}


@tf_keras.utils.register_keras_serializable(package='Vision')
class DilatedResNet(tf_keras.Model):
  """Creates a ResNet model with Deeplabv3 modifications.

  This backbone is suitable for semantic segmentation. This implements
    Liang-Chieh Chen, George Papandreou, Florian Schroff, Hartwig Adam.
    Rethinking Atrous Convolution for Semantic Image Segmentation.
    (https://arxiv.org/pdf/1706.05587)
  """

  def __init__(
      self,
      model_id: int,
      output_stride: int,
      input_specs: tf_keras.layers.InputSpec = layers.InputSpec(
          shape=[None, None, None, 3]),
      stem_type: str = 'v0',
      resnetd_shortcut: bool = False,
      replace_stem_max_pool: bool = False,
      se_ratio: Optional[float] = None,
      init_stochastic_depth_rate: float = 0.0,
      multigrid: Optional[Tuple[int]] = None,
      last_stage_repeats: int = 1,
      activation: str = 'relu',
      use_sync_bn: bool = False,
      norm_momentum: float = 0.99,
      norm_epsilon: float = 0.001,
      kernel_initializer: str = 'VarianceScaling',
      kernel_regularizer: Optional[tf_keras.regularizers.Regularizer] = None,
      bias_regularizer: Optional[tf_keras.regularizers.Regularizer] = None,
      **kwargs):
    """Initializes a ResNet model with DeepLab modification.

    Args:
      model_id: An `int` specifies depth of ResNet backbone model.
      output_stride: An `int` of output stride, ratio of input to output
        resolution.
      input_specs: A `tf_keras.layers.InputSpec` of the input tensor.
      stem_type: A `str` of stem type. Can be `v0` or `v1`. `v1` replaces 7x7
        conv by 3 3x3 convs.
      resnetd_shortcut: A `bool` of whether to use ResNet-D shortcut in
        downsampling blocks.
      replace_stem_max_pool: A `bool` of whether to replace the max pool in stem
        with a stride-2 conv,
      se_ratio: A `float` or None. Ratio of the Squeeze-and-Excitation layer.
      init_stochastic_depth_rate: A `float` of initial stochastic depth rate.
      multigrid: A tuple of the same length as the number of blocks in the last
        resnet stage.
      last_stage_repeats: An `int` that specifies how many times last stage is
        repeated.
      activation: A `str` name of the activation function.
      use_sync_bn: If True, use synchronized batch normalization.
      norm_momentum: A `float` of normalization momentum for the moving average.
      norm_epsilon: A `float` added to variance to avoid dividing by zero.
      kernel_initializer: A str for kernel initializer of convolutional layers.
      kernel_regularizer: A `tf_keras.regularizers.Regularizer` object for
        Conv2D. Default to None.
      bias_regularizer: A `tf_keras.regularizers.Regularizer` object for Conv2D.
        Default to None.
      **kwargs: Additional keyword arguments to be passed.
    """
    self._model_id = model_id
    self._output_stride = output_stride
    self._input_specs = input_specs
    self._use_sync_bn = use_sync_bn
    self._activation = activation
    self._norm_momentum = norm_momentum
    self._norm_epsilon = norm_epsilon
    self._norm = layers.BatchNormalization
    self._kernel_initializer = kernel_initializer
    self._kernel_regularizer = kernel_regularizer
    self._bias_regularizer = bias_regularizer
    self._stem_type = stem_type
    self._resnetd_shortcut = resnetd_shortcut
    self._replace_stem_max_pool = replace_stem_max_pool
    self._se_ratio = se_ratio
    self._init_stochastic_depth_rate = init_stochastic_depth_rate

    if tf_keras.backend.image_data_format() == 'channels_last':
      bn_axis = -1
    else:
      bn_axis = 1

    # Build ResNet.
    inputs = tf_keras.Input(shape=input_specs.shape[1:])

    if stem_type == 'v0':
      x = layers.Conv2D(
          filters=64,
          kernel_size=7,
          strides=2,
          use_bias=False,
          padding='same',
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer)(
              inputs)
      x = self._norm(
          axis=bn_axis,
          momentum=norm_momentum,
          epsilon=norm_epsilon,
          synchronized=use_sync_bn)(
              x)
      x = tf_utils.get_activation(activation)(x)
    elif stem_type == 'v1':
      x = layers.Conv2D(
          filters=64,
          kernel_size=3,
          strides=2,
          use_bias=False,
          padding='same',
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer)(
              inputs)
      x = self._norm(
          axis=bn_axis,
          momentum=norm_momentum,
          epsilon=norm_epsilon,
          synchronized=use_sync_bn)(
              x)
      x = tf_utils.get_activation(activation)(x)
      x = layers.Conv2D(
          filters=64,
          kernel_size=3,
          strides=1,
          use_bias=False,
          padding='same',
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer)(
              x)
      x = self._norm(
          axis=bn_axis,
          momentum=norm_momentum,
          epsilon=norm_epsilon,
          synchronized=use_sync_bn)(
              x)
      x = tf_utils.get_activation(activation)(x)
      x = layers.Conv2D(
          filters=128,
          kernel_size=3,
          strides=1,
          use_bias=False,
          padding='same',
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer)(
              x)
      x = self._norm(
          axis=bn_axis,
          momentum=norm_momentum,
          epsilon=norm_epsilon,
          synchronized=use_sync_bn)(
              x)
      x = tf_utils.get_activation(activation)(x)
    else:
      raise ValueError('Stem type {} not supported.'.format(stem_type))

    if replace_stem_max_pool:
      x = layers.Conv2D(
          filters=64,
          kernel_size=3,
          strides=2,
          use_bias=False,
          padding='same',
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer)(
              x)
      x = self._norm(
          axis=bn_axis,
          momentum=norm_momentum,
          epsilon=norm_epsilon,
          synchronized=use_sync_bn)(
              x)
      x = tf_utils.get_activation(activation, use_keras_layer=True)(x)
    else:
      x = layers.MaxPool2D(pool_size=3, strides=2, padding='same')(x)

    normal_resnet_stage = int(np.math.log2(self._output_stride)) - 2

    endpoints = {}
    for i in range(normal_resnet_stage + 1):
      spec = RESNET_SPECS[model_id][i]
      if spec[0] == 'bottleneck':
        block_fn = nn_blocks.BottleneckBlock
      else:
        raise ValueError('Block fn `{}` is not supported.'.format(spec[0]))
      x = self._block_group(
          inputs=x,
          filters=spec[1],
          strides=(1 if i == 0 else 2),
          dilation_rate=1,
          block_fn=block_fn,
          block_repeats=spec[2],
          stochastic_depth_drop_rate=nn_layers.get_stochastic_depth_rate(
              self._init_stochastic_depth_rate, i + 2, 4 + last_stage_repeats),
          name='block_group_l{}'.format(i + 2))
      endpoints[str(i + 2)] = x

    dilation_rate = 2
    for i in range(normal_resnet_stage + 1, 3 + last_stage_repeats):
      spec = RESNET_SPECS[model_id][i] if i < 3 else RESNET_SPECS[model_id][-1]
      if spec[0] == 'bottleneck':
        block_fn = nn_blocks.BottleneckBlock
      else:
        raise ValueError('Block fn `{}` is not supported.'.format(spec[0]))
      x = self._block_group(
          inputs=x,
          filters=spec[1],
          strides=1,
          dilation_rate=dilation_rate,
          block_fn=block_fn,
          block_repeats=spec[2],
          stochastic_depth_drop_rate=nn_layers.get_stochastic_depth_rate(
              self._init_stochastic_depth_rate, i + 2, 4 + last_stage_repeats),
          multigrid=multigrid if i >= 3 else None,
          name='block_group_l{}'.format(i + 2))
      dilation_rate *= 2

    endpoints[str(normal_resnet_stage + 2)] = x

    self._output_specs = {l: endpoints[l].get_shape() for l in endpoints}

    super(DilatedResNet, self).__init__(
        inputs=inputs, outputs=endpoints, **kwargs)

  def _block_group(self,
                   inputs: tf.Tensor,
                   filters: int,
                   strides: int,
                   dilation_rate: int,
                   block_fn: Callable[..., tf_keras.layers.Layer],
                   block_repeats: int = 1,
                   stochastic_depth_drop_rate: float = 0.0,
                   multigrid: Optional[List[int]] = None,
                   name: str = 'block_group'):
    """Creates one group of blocks for the ResNet model.

    Deeplab applies strides at the last block.

    Args:
      inputs: A `tf.Tensor` of size `[batch, channels, height, width]`.
      filters: An `int` off number of filters for the first convolution of the
        layer.
      strides: An `int` of stride to use for the first convolution of the layer.
        If greater than 1, this layer will downsample the input.
      dilation_rate: An `int` of diluted convolution rates.
      block_fn: Either `nn_blocks.ResidualBlock` or `nn_blocks.BottleneckBlock`.
      block_repeats: An `int` of number of blocks contained in the layer.
      stochastic_depth_drop_rate: A `float` of drop rate of the current block
        group.
      multigrid: A list of `int` or None. If specified, dilation rates for each
        block is scaled up by its corresponding factor in the multigrid.
      name: A `str` name for the block.

    Returns:
      The output `tf.Tensor` of the block layer.
    """
    if multigrid is not None and len(multigrid) != block_repeats:
      raise ValueError('multigrid has to match number of block_repeats')

    if multigrid is None:
      multigrid = [1] * block_repeats

    # TODO(arashwan): move striding at the of the block.
    x = block_fn(
        filters=filters,
        strides=strides,
        dilation_rate=dilation_rate * multigrid[0],
        use_projection=True,
        stochastic_depth_drop_rate=stochastic_depth_drop_rate,
        se_ratio=self._se_ratio,
        resnetd_shortcut=self._resnetd_shortcut,
        kernel_initializer=self._kernel_initializer,
        kernel_regularizer=self._kernel_regularizer,
        bias_regularizer=self._bias_regularizer,
        activation=self._activation,
        use_sync_bn=self._use_sync_bn,
        norm_momentum=self._norm_momentum,
        norm_epsilon=self._norm_epsilon)(
            inputs)
    for i in range(1, block_repeats):
      x = block_fn(
          filters=filters,
          strides=1,
          dilation_rate=dilation_rate * multigrid[i],
          use_projection=False,
          stochastic_depth_drop_rate=stochastic_depth_drop_rate,
          resnetd_shortcut=self._resnetd_shortcut,
          se_ratio=self._se_ratio,
          kernel_initializer=self._kernel_initializer,
          kernel_regularizer=self._kernel_regularizer,
          bias_regularizer=self._bias_regularizer,
          activation=self._activation,
          use_sync_bn=self._use_sync_bn,
          norm_momentum=self._norm_momentum,
          norm_epsilon=self._norm_epsilon)(
              x)

    return tf.identity(x, name=name)

  def get_config(self):
    config_dict = {
        'model_id': self._model_id,
        'output_stride': self._output_stride,
        'stem_type': self._stem_type,
        'resnetd_shortcut': self._resnetd_shortcut,
        'replace_stem_max_pool': self._replace_stem_max_pool,
        'se_ratio': self._se_ratio,
        'init_stochastic_depth_rate': self._init_stochastic_depth_rate,
        'activation': self._activation,
        'use_sync_bn': self._use_sync_bn,
        'norm_momentum': self._norm_momentum,
        'norm_epsilon': self._norm_epsilon,
        'kernel_initializer': self._kernel_initializer,
        'kernel_regularizer': self._kernel_regularizer,
        'bias_regularizer': self._bias_regularizer,
    }
    return config_dict

  @classmethod
  def from_config(cls, config, custom_objects=None):
    return cls(**config)

  @property
  def output_specs(self):
    """A dict of {level: TensorShape} pairs for the model output."""
    return self._output_specs


@factory.register_backbone_builder('dilated_resnet')
def build_dilated_resnet(
    input_specs: tf_keras.layers.InputSpec,
    backbone_config: hyperparams.Config,
    norm_activation_config: hyperparams.Config,
    l2_regularizer: tf_keras.regularizers.Regularizer = None) -> tf_keras.Model:  # pytype: disable=annotation-type-mismatch  # typed-keras
  """Builds ResNet backbone from a config."""
  backbone_type = backbone_config.type
  backbone_cfg = backbone_config.get()
  assert backbone_type == 'dilated_resnet', (f'Inconsistent backbone type '
                                             f'{backbone_type}')

  return DilatedResNet(
      model_id=backbone_cfg.model_id,
      output_stride=backbone_cfg.output_stride,
      input_specs=input_specs,
      stem_type=backbone_cfg.stem_type,
      resnetd_shortcut=backbone_cfg.resnetd_shortcut,
      replace_stem_max_pool=backbone_cfg.replace_stem_max_pool,
      se_ratio=backbone_cfg.se_ratio,
      init_stochastic_depth_rate=backbone_cfg.stochastic_depth_drop_rate,
      multigrid=backbone_cfg.multigrid,
      last_stage_repeats=backbone_cfg.last_stage_repeats,
      activation=norm_activation_config.activation,
      use_sync_bn=norm_activation_config.use_sync_bn,
      norm_momentum=norm_activation_config.norm_momentum,
      norm_epsilon=norm_activation_config.norm_epsilon,
      kernel_regularizer=l2_regularizer)