Compressed Screen Content Image Super Resolution

Screen content has become one of the prominent mediums in increasingly connected world. With prevalence remote collaboration and communication such as virtual conferences online education, recent years have witnessed a dramatic increase data volume screen content. compression serves fundamental technology fostering storage, transmission, exhibition In this article, we target super-resolution compressed images, intending to tackle real-world challenge problems. A dataset is proposed for contents contaminated with different distortion levels. Subsequently, introduce principle multi-hypothesis into super resolution propose new paradigm restoration images. The luminance sharpness similarity metric adopted network learning better adapt characteristic ensure perceptual fidelity. Experimental results verify superiority effectiveness method.