Copy-and-Paste Network¶
Module containing the implementation of the Copy-and-Paste Network (CPN). This implementation has been slightly modified to fill the requirements of this thesis. The original version can be found in:
https://github.com/shleecs/Copy-and-Paste-Networks-for-Deep-Video-Inpainting
- class master_thesis.model_cpn.CPN¶
Bases:
torch.nn.modules.module.Module
Implementation of the Copy-and-Paste Network (CPN).
- forward(x_target, m_target, x_refs, m_refs)¶
Forward pass through the Copy-and-Paste Network (CPN).
- Parameters
x_target –
m_target –
x_refs –
m_refs –
Returns:
- align(x_target, m_target, x_refs, m_refs)¶
- copy_and_paste(x_target, v_target, x_aligned, v_aligned)¶
- inpaint(x, m)¶
- static get_indexes(t, n_frames, p=2, r_list_max_length=120)¶
- static init_He(module)¶
- static init_model_with_state(checkpoint_path, device='cpu')¶
- training: bool¶
- class master_thesis.model_cpn.A_Encoder¶
Bases:
torch.nn.modules.module.Module
- forward(in_f, in_v)¶
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 master_thesis.model_cpn.A_Regressor¶
Bases:
torch.nn.modules.module.Module
- forward(feat1, feat2)¶
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 master_thesis.model_cpn.Encoder¶
Bases:
torch.nn.modules.module.Module
- forward(in_f, in_v)¶
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 master_thesis.model_cpn.CM_Module¶
Bases:
torch.nn.modules.module.Module
- forward(c_feats, v_t, v_aligned)¶
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.
- static masked_softmax(vec, mask, dim)¶
- training: bool¶
- class master_thesis.model_cpn.Decoder¶
Bases:
torch.nn.modules.module.Module
- forward(x)¶
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 master_thesis.model_cpn.Conv2d(in_ch, out_ch, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), D=(1, 1), activation=None)¶
Bases:
torch.nn.modules.module.Module
- forward(x)¶
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¶