Source code for torchvision.models.segmentation.fcn
from functools import partial
from typing import Any, Optional
from torch import nn
from ...transforms._presets import SemanticSegmentation
from .._api import register_model, Weights, WeightsEnum
from .._meta import _VOC_CATEGORIES
from .._utils import _ovewrite_value_param, handle_legacy_interface, IntermediateLayerGetter
from ..resnet import ResNet, resnet101, ResNet101_Weights, resnet50, ResNet50_Weights
from ._utils import _SimpleSegmentationModel
__all__ = ["FCN", "FCN_ResNet50_Weights", "FCN_ResNet101_Weights", "fcn_resnet50", "fcn_resnet101"]
class FCN(_SimpleSegmentationModel):
"""
Implements FCN model from
`"Fully Convolutional Networks for Semantic Segmentation"
<https://arxiv.org/abs/1411.4038>`_.
Args:
backbone (nn.Module): the network used to compute the features for the model.
The backbone should return an OrderedDict[Tensor], with the key being
"out" for the last feature map used, and "aux" if an auxiliary classifier
is used.
classifier (nn.Module): module that takes the "out" element returned from
the backbone and returns a dense prediction.
aux_classifier (nn.Module, optional): auxiliary classifier used during training
"""
pass
class FCNHead(nn.Sequential):
def __init__(self, in_channels: int, channels: int) -> None:
inter_channels = in_channels // 4
layers = [
nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False),
nn.BatchNorm2d(inter_channels),
nn.ReLU(),
nn.Dropout(0.1),
nn.Conv2d(inter_channels, channels, 1),
]
super().__init__(*layers)
_COMMON_META = {
"categories": _VOC_CATEGORIES,
"min_size": (1, 1),
"_docs": """
These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
dataset.
""",
}
[docs]class FCN_ResNet50_Weights(WeightsEnum):
COCO_WITH_VOC_LABELS_V1 = Weights(
url="https://download.pytorch.org/models/fcn_resnet50_coco-1167a1af.pth",
transforms=partial(SemanticSegmentation, resize_size=520),
meta={
**_COMMON_META,
"num_params": 35322218,
"recipe": "https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50",
"_metrics": {
"COCO-val2017-VOC-labels": {
"miou": 60.5,
"pixel_acc": 91.4,
}
},
},
)
DEFAULT = COCO_WITH_VOC_LABELS_V1
[docs]class FCN_ResNet101_Weights(WeightsEnum):
COCO_WITH_VOC_LABELS_V1 = Weights(
url="https://download.pytorch.org/models/fcn_resnet101_coco-7ecb50ca.pth",
transforms=partial(SemanticSegmentation, resize_size=520),
meta={
**_COMMON_META,
"num_params": 54314346,
"recipe": "https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101",
"_metrics": {
"COCO-val2017-VOC-labels": {
"miou": 63.7,
"pixel_acc": 91.9,
}
},
},
)
DEFAULT = COCO_WITH_VOC_LABELS_V1
def _fcn_resnet(
backbone: ResNet,
num_classes: int,
aux: Optional[bool],
) -> FCN:
return_layers = {"layer4": "out"}
if aux:
return_layers["layer3"] = "aux"
backbone = IntermediateLayerGetter(backbone, return_layers=return_layers)
aux_classifier = FCNHead(1024, num_classes) if aux else None
classifier = FCNHead(2048, num_classes)
return FCN(backbone, classifier, aux_classifier)
[docs]@register_model()
@handle_legacy_interface(
weights=("pretrained", FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1),
weights_backbone=("pretrained_backbone", ResNet50_Weights.IMAGENET1K_V1),
)
def fcn_resnet50(
*,
weights: Optional[FCN_ResNet50_Weights] = None,
progress: bool = True,
num_classes: Optional[int] = None,
aux_loss: Optional[bool] = None,
weights_backbone: Optional[ResNet50_Weights] = ResNet50_Weights.IMAGENET1K_V1,
**kwargs: Any,
) -> FCN:
"""Fully-Convolutional Network model with a ResNet-50 backbone from the `Fully Convolutional
Networks for Semantic Segmentation <https://arxiv.org/abs/1411.4038>`_ paper.
.. betastatus:: segmentation module
Args:
weights (:class:`~torchvision.models.segmentation.FCN_ResNet50_Weights`, optional): The
pretrained weights to use. See
:class:`~torchvision.models.segmentation.FCN_ResNet50_Weights` below for
more details, and possible values. By default, no pre-trained
weights are used.
progress (bool, optional): If True, displays a progress bar of the
download to stderr. Default is True.
num_classes (int, optional): number of output classes of the model (including the background).
aux_loss (bool, optional): If True, it uses an auxiliary loss.
weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained
weights for the backbone.
**kwargs: parameters passed to the ``torchvision.models.segmentation.fcn.FCN``
base class. Please refer to the `source code
<https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py>`_
for more details about this class.
.. autoclass:: torchvision.models.segmentation.FCN_ResNet50_Weights
:members:
"""
weights = FCN_ResNet50_Weights.verify(weights)
weights_backbone = ResNet50_Weights.verify(weights_backbone)
if weights is not None:
weights_backbone = None
num_classes = _ovewrite_value_param("num_classes", num_classes, len(weights.meta["categories"]))
aux_loss = _ovewrite_value_param("aux_loss", aux_loss, True)
elif num_classes is None:
num_classes = 21
backbone = resnet50(weights=weights_backbone, replace_stride_with_dilation=[False, True, True])
model = _fcn_resnet(backbone, num_classes, aux_loss)
if weights is not None:
model.load_state_dict(weights.get_state_dict(progress=progress))
return model
[docs]@register_model()
@handle_legacy_interface(
weights=("pretrained", FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1),
weights_backbone=("pretrained_backbone", ResNet101_Weights.IMAGENET1K_V1),
)
def fcn_resnet101(
*,
weights: Optional[FCN_ResNet101_Weights] = None,
progress: bool = True,
num_classes: Optional[int] = None,
aux_loss: Optional[bool] = None,
weights_backbone: Optional[ResNet101_Weights] = ResNet101_Weights.IMAGENET1K_V1,
**kwargs: Any,
) -> FCN:
"""Fully-Convolutional Network model with a ResNet-101 backbone from the `Fully Convolutional
Networks for Semantic Segmentation <https://arxiv.org/abs/1411.4038>`_ paper.
.. betastatus:: segmentation module
Args:
weights (:class:`~torchvision.models.segmentation.FCN_ResNet101_Weights`, optional): The
pretrained weights to use. See
:class:`~torchvision.models.segmentation.FCN_ResNet101_Weights` below for
more details, and possible values. By default, no pre-trained
weights are used.
progress (bool, optional): If True, displays a progress bar of the
download to stderr. Default is True.
num_classes (int, optional): number of output classes of the model (including the background).
aux_loss (bool, optional): If True, it uses an auxiliary loss.
weights_backbone (:class:`~torchvision.models.ResNet101_Weights`, optional): The pretrained
weights for the backbone.
**kwargs: parameters passed to the ``torchvision.models.segmentation.fcn.FCN``
base class. Please refer to the `source code
<https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py>`_
for more details about this class.
.. autoclass:: torchvision.models.segmentation.FCN_ResNet101_Weights
:members:
"""
weights = FCN_ResNet101_Weights.verify(weights)
weights_backbone = ResNet101_Weights.verify(weights_backbone)
if weights is not None:
weights_backbone = None
num_classes = _ovewrite_value_param("num_classes", num_classes, len(weights.meta["categories"]))
aux_loss = _ovewrite_value_param("aux_loss", aux_loss, True)
elif num_classes is None:
num_classes = 21
backbone = resnet101(weights=weights_backbone, replace_stride_with_dilation=[False, True, True])
model = _fcn_resnet(backbone, num_classes, aux_loss)
if weights is not None:
model.load_state_dict(weights.get_state_dict(progress=progress))
return model
# The dictionary below is internal implementation detail and will be removed in v0.15
from .._utils import _ModelURLs
model_urls = _ModelURLs(
{
"fcn_resnet50_coco": FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1.url,
"fcn_resnet101_coco": FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1.url,
}
)