Advanced Super-Resolution using Lossless Pooling Convolutional Networks
This work addresses the challenge of achieving high-quality image upscaling for applications like photography and video processing, representing an incremental advancement over existing methods.
The paper tackles the problem of still image super-resolution by introducing a novel deep learning approach that uses artificially created self-replicas of the input image to enhance upsampling, resulting in significant quality improvements as confirmed by evaluations.
In this paper, we present a novel deep learning-based approach for still image super-resolution, that unlike the mainstream models does not rely solely on the input low resolution image for high quality upsampling, and takes advantage of a set of artificially created auxiliary self-replicas of the input image that are incorporated in the neural network to create an enhanced and accurate upscaling scheme. Inclusion of the proposed lossless pooling layers, and the fusion of the input self-replicas enable the model to exploit the high correlation between multiple instances of the same content, and eventually result in significant improvements in the quality of the super-resolution, which is confirmed by extensive evaluations.