diffrp.utils.composite#

diffrp.utils.composite.additive(bgc: Tensor, bga: Tensor, fgc: Tensor, fga: Tensor)#

diffrp.utils.composite.alpha_additive(bgc: Tensor, bga: Tensor, fgc: Tensor, fga: Tensor)#

diffrp.utils.composite.alpha_blend(bgc: Tensor, bga: Tensor, fgc: Tensor, fga: Tensor)#

diffrp.utils.composite.background_alpha_compose(bg: Tensor | List[float | int] | float | int, fgca: Tensor)#

Compose a background color to a foreground via alpha blending.

It not only applies to RGB color and RGBA foregrounds, but also general linear data.

The background should be one element shorter than the foreground elements.

Parameters:

bg – Tensor or List[float] or number. A single number is equivalent to the RGB list filled by it. The length should be one less than foreground.
fgca – Tensor with alpha, shape (…, C + 1).

Returns:

Composed result, shape (…, C).

Return type:

torch.Tensor