Multi-channel convolutional neural networks for image super-resolution

Abstract

This paper proposes image super-resolution techniques with multi-channel convolutional neural networks. In the proposed method, output pixels are classified into K × K groups depending on their coordinates. Those groups are generated from separate channels of a convolutional neural network (CNN). Finally, they are synthesized into a K × K magnified image. This architecture can enlarge images directly without bicubic interpolation. Experimental results of 2×2, 3×3, and 4×4 magnifications have shown that the average PSNR for the proposed method is about 0.2 dB higher than that for the conventional SRCNN.