ITSRN++: Stronger and better implicit transformer network for screen content image continuous super-resolution

Sheng Shen, Huanjing Yue, Kun Li, Jingyu Yang

Abstract: Highlights•A novel modulation-based transformer upsampler enhances screen content image super-resolution.•Dual Branch Block (DBB) architecture effectively captures both high- and low-frequency details.•The newly introduced SCI2K dataset facilitates advancements in screen content image research.•ITSRN++ outperforms existing methods in both continuous and discrete super-resolution tasks.•The proposed upsampler and DBB can be integrated into other SR methods to boost performance.