Multi-Grid Back-Projection Networks
This work provides an efficient and effective architecture for image and video restoration, particularly beneficial for applications requiring high-resolution content from low-resolution inputs, such as in media processing and computer vision.
The paper introduces Multi-Grid Back-Projection (MGBP), a fully-convolutional network architecture designed for image and video restoration from upscaling artifacts. MGBP achieves performance comparable to state-of-the-art methods for exact high-resolution recovery and offers a novel strategy using multi-scale noise inputs to control artificial detail generation for perceptual quality targets.
Multi-Grid Back-Projection (MGBP) is a fully-convolutional network architecture that can learn to restore images and videos with upscaling artifacts. Using the same strategy of multi-grid partial differential equation (PDE) solvers this multiscale architecture scales computational complexity efficiently with increasing output resolutions. The basic processing block is inspired in the iterative back-projection (IBP) algorithm and constitutes a type of cross-scale residual block with feedback from low resolution references. The architecture performs in par with state-of-the-arts alternatives for regression targets that aim to recover an exact copy of a high resolution image or video from which only a downscale image is known. A perceptual quality target aims to create more realistic outputs by introducing artificial changes that can be different from a high resolution original content as long as they are consistent with the low resolution input. For this target we propose a strategy using noise inputs in different resolution scales to control the amount of artificial details generated in the output. The noise input controls the amount of innovation that the network uses to create artificial realistic details. The effectiveness of this strategy is shown in benchmarks and it is explained as a particular strategy to traverse the perception-distortion plane.