search

dvlab-research / SNR-Aware-Low-Light-Enhance

This is the official implementation for the paper "SNR-aware low-light image enhancement" in CVPR2022
171 stars 19 forks source link

Confusion about equation 6 in the paper #3

Closed biubiubiiu closed 2 years ago

biubiubiiu commented 2 years ago

Equation 6 is the paper is

image

I'm confused by the element-wise addition here. Since the SNR map is used to mask out tokens (i.e. patches) with low SNR, shouldn't the element-wise add be the element-wise product here? I.e.:

$$ \operatorname{Softmax}\left(\frac{\mathbf{Q}{i, b} \mathbf{K}^T{i, b}}{\sqrt{db}} \circ \left(1-\mathcal{S}^{\prime}\right) \sigma\right) \mathbf{V}{i, b} $$

https://github.com/dvlab-research/SNR-Aware-Low-Light-Enhance/blob/d55031160422be092ec526c465654c436e12a2b5/models/archs/transformer/Modules.py#L15-L24

xiaogang00 commented 2 years ago

As indicated in the paper, sigma is a negative scalar, with a large absolute value. This number is added for the attention map with the mask map. And for the area which is added with sigma, its value after softmax will be zero, meaning that these areas (these areas normally have low SNR) will not be included in the attention computation.

biubiubiiu commented 2 years ago

Thank you for your clarification.

As indicated in the paper, sigma is a negative scalar, with a large absolute value. This number is added for the attention map with the mask map. And for the area which is added with sigma, its value after softmax will be zero, meaning that these areas (these areas normally have low SNR) will not be included in the attention computation.