mmagic.models.editors.gca.gca_module
¶
Module Contents¶
Classes¶
Guided Contextual Attention Module. |
- class mmagic.models.editors.gca.gca_module.GCAModule(in_channels, out_channels, kernel_size=3, stride=1, rate=2, pad_args=dict(mode='reflect'), interpolation='nearest', penalty=- 10000.0, eps=0.0001)[源代码]¶
Bases:
torch.nn.Module
Guided Contextual Attention Module.
From https://arxiv.org/pdf/2001.04069.pdf. Based on https://github.com/nbei/Deep-Flow-Guided-Video-Inpainting. This module use image feature map to augment the alpha feature map with guided contextual attention score.
Image feature and alpha feature are unfolded to small patches and later used as conv kernel. Thus, we refer the unfolding size as kernel size. Image feature patches have a default kernel size 3 while the kernel size of alpha feature patches could be specified by rate (see rate below). The image feature patches are used to convolve with the image feature itself to calculate the contextual attention. Then the attention feature map is convolved by alpha feature patches to obtain the attention alpha feature. At last, the attention alpha feature is added to the input alpha feature.
- 参数
in_channels (int) – Input channels of the guided contextual attention module.
out_channels (int) – Output channels of the guided contextual attention module.
kernel_size (int) – Kernel size of image feature patches. Default 3.
stride (int) – Stride when unfolding the image feature. Default 1.
rate (int) – The downsample rate of image feature map. The corresponding kernel size and stride of alpha feature patches will be rate x 2 and rate. It could be regarded as the granularity of the gca module. Default: 2.
pad_args (dict) – Parameters of padding when convolve image feature with image feature patches or alpha feature patches. Allowed keys are mode and value. See torch.nn.functional.pad() for more information. Default: dict(mode=’reflect’).
interpolation (str) – Interpolation method in upsampling and downsampling.
penalty (float) – Punishment hyperparameter to avoid a large correlation between each unknown patch and itself. Default: -1e4.
eps (float) – A small number to avoid dividing by 0 when calculating the normed image feature patch. Default: 1e-4.
- forward(img_feat, alpha_feat, unknown=None, softmax_scale=1.0)[源代码]¶
Forward function of GCAModule.
- 参数
img_feat (Tensor) – Image feature map of shape (N, ori_c, ori_h, ori_w).
alpha_feat (Tensor) – Alpha feature map of shape (N, alpha_c, ori_h, ori_w).
unknown (Tensor, optional) – Unknown area map generated by trimap. If specified, this tensor should have shape (N, 1, ori_h, ori_w).
softmax_scale (float, optional) – The softmax scale of the attention if unknown area is not provided in forward. Default: 1.
- 返回
The augmented alpha feature.
- 返回类型
Tensor
- extract_feature_maps_patches(img_feat, alpha_feat, unknown)[源代码]¶
Extract image feature, alpha feature unknown patches.
- 参数
img_feat (Tensor) – Image feature map of shape (N, img_c, img_h, img_w).
alpha_feat (Tensor) – Alpha feature map of shape (N, alpha_c, ori_h, ori_w).
unknown (Tensor, optional) – Unknown area map generated by trimap of shape (N, 1, img_h, img_w).
- 返回
3-tuple of
Tensor
: Image feature patches of shape (N, img_h*img_w, img_c, img_ks, img_ks).Tensor
: Guided contextual attention alpha feature map. (N, img_h*img_w, alpha_c, alpha_ks, alpha_ks).Tensor
: Unknown mask of shape (N, img_h*img_w, 1, 1).- 返回类型
tuple
- compute_similarity_map(img_feat, img_ps)[源代码]¶
Compute similarity between image feature patches.
- 参数
img_feat (Tensor) – Image feature map of shape (1, img_c, img_h, img_w).
img_ps (Tensor) – Image feature patches tensor of shape (1, img_h*img_w, img_c, img_ks, img_ks).
- 返回
Similarity map between image feature patches with shape (1, img_h*img_w, img_h, img_w).
- 返回类型
Tensor
- compute_guided_attention_score(similarity_map, unknown_ps, scale, self_mask)[源代码]¶
Compute guided attention score.
- 参数
similarity_map (Tensor) – Similarity map of image feature with shape (1, img_h*img_w, img_h, img_w).
unknown_ps (Tensor) – Unknown area patches tensor of shape (1, img_h*img_w, 1, 1).
scale (Tensor) – Softmax scale of known and unknown area: [unknown_scale, known_scale].
self_mask (Tensor) – Self correlation mask of shape (1, img_h*img_w, img_h, img_w). At (1, i*i, i, i) mask value equals -1e4 for i in [1, img_h*img_w] and other area is all zero.
- 返回
Similarity map between image feature patches with shape (1, img_h*img_w, img_h, img_w).
- 返回类型
Tensor
- propagate_alpha_feature(gca_score, alpha_ps)[源代码]¶
Propagate alpha feature based on guided attention score.
- 参数
gca_score (Tensor) – Guided attention score map of shape (1, img_h*img_w, img_h, img_w).
alpha_ps (Tensor) – Alpha feature patches tensor of shape (1, img_h*img_w, alpha_c, alpha_ks, alpha_ks).
- 返回
Propagated alpha feature map of shape (1, alpha_c, alpha_h, alpha_w).
- 返回类型
Tensor
- process_unknown_mask(unknown, img_feat, softmax_scale)[源代码]¶
Process unknown mask.
- 参数
unknown (Tensor, optional) – Unknown area map generated by trimap of shape (N, 1, ori_h, ori_w)
img_feat (Tensor) – The interpolated image feature map of shape (N, img_c, img_h, img_w).
softmax_scale (float, optional) – The softmax scale of the attention if unknown area is not provided in forward. Default: 1.
- 返回
2-tuple of
Tensor
: Interpolated unknown area map of shape (N, img_h*img_w, img_h, img_w).Tensor
: Softmax scale tensor of known and unknown area of shape (N, 2).- 返回类型
tuple
- extract_patches(x, kernel_size, stride)[源代码]¶
Extract feature patches.
The feature map will be padded automatically to make sure the number of patches is equal to (H / stride) * (W / stride).
- 参数
x (Tensor) – Feature map of shape (N, C, H, W).
kernel_size (int) – Size of each patches.
stride (int) – Stride between patches.
- 返回
Extracted patches of shape (N, (H / stride) * (W / stride) , C, kernel_size, kernel_size).
- 返回类型
Tensor
- pad(x, kernel_size, stride)[源代码]¶
Pad input tensor.
- 参数
x (Tensor) – Input tensor.
kernel_size (int) – Kernel size of conv layer.
stride (int) – Stride of conv layer.
- 返回
Padded tensor
- 返回类型
Tensor