… The proper and adequate usage of such information can enhance the results of video
super-resolution. Therefore, we build a taxonomy for existing video super-resolution methods …

… to image super-resolution (SR) as … video super-resolution. We propose a CNN that is trained
on both the spatial and the temporal dimensions of videos to enhance their spatial resolution…

… with strong single image and video superresolution baselines using identical neural networks
… Finally, we also compare FRVSR with several existing video super-resolution approaches …

… of video superresolution has drawn much attention in both the research and industrial
communities. For example, video super-resolution is required when videos … , or when videos are …

Previous CNN-based video super-resolution approaches need to align multiple frames to the
reference. In this paper, we show that proper frame alignment and motion compensation is …

Video super-resolution (VSR) approaches tend to have more components than the image
counterparts as they need to exploit the additional temporal dimension. Complex designs are …

… We also evaluated run time of 1080 HD video superresolution using videos from the Xiph
and the Ultra Video Group database. With upscale factor of 3, SRCNN 9-5-5 ImageNet model …

… of video super-resolution. The state-of-theart method BasicVSR adopts bidirectional propagation
with feature alignment to effectively exploit information from the entire input video. … video …

… the feature propagation in video frames which contains rich … and align such information in
a video sequence. However, most … a new Transformer for video super-resolution, called VSR-…

… Video super-resolution (SR) is the task of inferring a high-resolution (HR) video sequence
from a low-resolution … this problem is challenging because video signals vary in both temporal …