Abstract: We propose reparameterized refocusing convolution (RefConv) as a replacement for regular convolutional layers, which is a plug-and-play module to improve the performance without any ...
Abstract: Pixel-level adaptive convolution, which overcomes the deficiency of the spatial-invariance of standard convolution, is always limited to performing feature extraction from local patches and ...
In this paper, we tackle the high computational overhead of transformers for lightweight image super-resolution. (SR). Motivated by the observations of self-attention's inter-layer repetition, we ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results