Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Images
This work addresses the need for efficient graphics super-resolution in industry and research communities, offering a domain-specific incremental improvement by optimizing for rasterized images.
The paper tackles the problem of high computational cost in rendering high-resolution graphics by introducing Super-Resolution by Predicting Offsets (SRPO), an ultra-efficient network that processes rasterized images to achieve real-time super-resolution with only 8,434 parameters, resulting in superior visual effects at a smaller computational cost compared to state-of-the-art methods.
Rendering high-resolution (HR) graphics brings substantial computational costs. Efficient graphics super-resolution (SR) methods may achieve HR rendering with small computing resources and have attracted extensive research interests in industry and research communities. We present a new method for real-time SR for computer graphics, namely Super-Resolution by Predicting Offsets (SRPO). Our algorithm divides the image into two parts for processing, i.e., sharp edges and flatter areas. For edges, different from the previous SR methods that take the anti-aliased images as inputs, our proposed SRPO takes advantage of the characteristics of rasterized images to conduct SR on the rasterized images. To complement the residual between HR and low-resolution (LR) rasterized images, we train an ultra-efficient network to predict the offset maps to move the appropriate surrounding pixels to the new positions. For flat areas, we found simple interpolation methods can already generate reasonable output. We finally use a guided fusion operation to integrate the sharp edges generated by the network and flat areas by the interpolation method to get the final SR image. The proposed network only contains 8,434 parameters and can be accelerated by network quantization. Extensive experiments show that the proposed SRPO can achieve superior visual effects at a smaller computational cost than the existing state-of-the-art methods.