Recently, learning-based models have enhanced the performance of Single-Image Super-Resolution (SISR). However, applying SISR successively to each video frame leads to lack of temporal consistency. On ...