Fig. 3
From: Diffusion-based heterogeneous network for ancient mural restoration
The architecture of the proposed pixel space augmentation block. The input feature x is processed through parallel convolutional paths. One path extracts contextual features, which are upsampled and passed through a sigmoid activation to generate an attention map m. The final output \({\tilde{x}}\) is obtained via element-wise multiplication of x and m, as denoted by the circled dot symbol.
