bob.ip.binseg.models.hed

Functions

hed([pretrained_backbone, progress])

Builds HED by adding backbone and head together

Classes

ConcatFuseBlock()	Takes in five feature maps with one channel each, concatenates thems and applies a 1x1 convolution with 1 output channel.
HED([in_channels_list])	HED head module

class bob.ip.binseg.models.hed.ConcatFuseBlock

Bases: torch.nn.modules.module.Module

Takes in five feature maps with one channel each, concatenates thems and applies a 1x1 convolution with 1 output channel.

forward(x1, x2, x3, x4, x5)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class bob.ip.binseg.models.hed.HED(in_channels_list=None)

Bases: torch.nn.modules.module.Module

HED head module

Parameters

in_channels_list (list) – number of channels for each feature map that is returned from backbone

forward(x)

Parameters

x (list) – list of tensors as returned from the backbone network. First element: height and width of input image. Remaining elements: feature maps for each feature level.

Returns

tensor

Return type

torch.Tensor

training: bool

bob.ip.binseg.models.hed.hed(pretrained_backbone=True, progress=True)

Builds HED by adding backbone and head together

Parameters

• pretrained_backbone (bool, Optional) – If set to True, then loads a pre-trained version of the backbone (not the head) for the DRIU network using VGG-16 trained for ImageNet classification.

• progress (bool, Optional) – If set to True, and you decided to use a pretrained_backbone, then, shows a progress bar of the backbone model downloading if download is necesssary.

Returns

module – Network model for HED

Return type

torch.nn.Module