ClassSR

ClassSR

ClassSR: A Framework for Accelerated Super-Resolution Networks ClassSR is a framework designed to accelerate super-resolution (SR) networks on large images ranging from 2K to 8K. It combines classification and SR within a unified framework. The framework first utilizes a Class-Module to classify sub-images into different classes based on restoration difficulties. Then, it applies an SR-Module to perform SR for the different classes. The Class-Module uses a conventional classification network, w

PULSE

PULSE

Overview of PULSE Algorithm If you love taking photos, then you know how frustrating it can be when your favorite shot turns out blurry or low-quality. Fortunately, researchers have come up with a solution for this problem, known as PULSE. This innovative algorithm allows you to enhance the resolution of your photos while maintaining their natural look and feel. The PULSE algorithm works by using a technique called self-supervised photo upsampling. Rather than simply adding detail to a low-res

1 / 1