CVPR2025
VideoGigaGAN: Towards Detail-rich Video Super-Resolution
Yiran Xu, Taesung Park, Richard Zhang, Yang Zhou, Eli Shechtman, Feng Liu, Jia-Bin Huang, Difan Liu
Abstract
Figure 1 . We present VideoGigaGAN, a generative video super-resolution model that upscales videos with high-frequency details while preserving temporal consistency. Top: We compare our approach with TTVSR [39] and BasicVSR++ [8]. Our method produces better temporal consistency and finer details than previous methods. Bottom: Our model produces high-quality videos with 8× super-resolution.