ASDN: A Deep Convolutional Network for Arbitrary Scale Image Super-Resolution
This addresses the need for efficient super-resolution in image viewer applications where users zoom to arbitrary scales, offering a computationally efficient solution with incremental improvements over existing methods.
The paper tackles the problem of arbitrary scale image super-resolution, which is computationally expensive due to requiring training at many scales, by proposing ASDN, a deep convolutional network that uses a Laplacian pyramid method to reconstruct high-resolution images efficiently, achieving about 1 dB higher PSNR than predefined methods at fixed scales and generally exceeding Meta-SR at arbitrary scales.
Deep convolutional neural networks have significantly improved the peak signal-to-noise ratio of SuperResolution (SR). However, image viewer applications commonly allow users to zoom the images to arbitrary magnification scales, thus far imposing a large number of required training scales at a tremendous computational cost. To obtain a more computationally efficient model for arbitrary scale SR, this paper employs a Laplacian pyramid method to reconstruct any-scale high-resolution (HR) images using the high-frequency image details in a Laplacian Frequency Representation. For SR of small-scales (between 1 and 2), images are constructed by interpolation from a sparse set of precalculated Laplacian pyramid levels. SR of larger scales is computed by recursion from small scales, which significantly reduces the computational cost. For a full comparison, fixed- and any-scale experiments are conducted using various benchmarks. At fixed scales, ASDN outperforms predefined upsampling methods (e.g., SRCNN, VDSR, DRRN) by about 1 dB in PSNR. At any-scale, ASDN generally exceeds Meta-SR on many scales.