Abstract In this paper, we are the first to propose a new graph convolution-based decoder namely, Cascaded Graph Con-volutional Attention Decoder (G-CASCADE), for 2D med-ical image …

2.2. Re-parameterization ion has become an effective technique for eficient neural network design. Lei et al. [6] proposed an asymmetric convolution block (ACB) to s rengthen the vanilla …

Multiplying by a circulant matrix is equivalent to a very famous operation called a circular convolution. Convolution operations, and hence circulant matrices, show up in lots of …

tion mechanism: focusing on important features and sup-pressing unnecessary ones. In this paper, e propose a new network module, named “Convolutional Block Attention Module”. …

In this paper, we propose a new network module, named “Convolutional Block Attention Module”. Since convolution operations extract informative features by blending cross-channel and …

A conformer block is composed of four modules stacked together, i.e, a feed-forward module, a self-attention module, a convolution module, and a second feed-forward module in the end. …

We design a more suitable re-parameterization block, namely Edge-oriented Convolution Block (ECB), which can more efectively extract edge and texture information for the SR task.