BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment

A recurrent structure is a popular framework choice for the task of video super-resolution. The state-of-the-art method BasicVSR adopts bidirectional propagation with feature alignment to effectively exploit information from the entire input video. In this study, we redesign BasicVSR by proposing second-order grid propagation and flow-guided deformable alignment. We show that by empowering the recurrent framework with the enhanced propagation and alignment, one can exploit spatiotemporal information across misaligned video frames more effectively. The new components lead to an improved performance under a similar computational constraint. In particular, our model BasicVSR++ surpasses BasicVSR by 0.82 dB in PSNR with similar number of parameters. In addition to video super-resolution, BasicVSR++ generalizes well to other video restoration tasks such as compressed video enhancement. In NTIRE 2021, BasicVSR++ obtains three champions and one runner-up in the Video Super-Resolution and Compressed Video Enhancement Challenges. Codes and models will be released to MMEditing.

PDF Abstract CVPR 2022 PDF CVPR 2022 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Video Enhancement MFQE v2 BasicVSR++ Incremental PSNR 1.10 # 1
Video Super-Resolution MSU Video Upscalers: Quality Enhancement BasicVsr++RD PSNR 30.98 # 8
LPIPS 0.334 # 28
SSIM 0.881 # 28
Analog Video Restoration TAPE BasicVSR++ LPIPS 0.098 # 3
VMAF 78.91 # 2
PSNR 31.66 # 5
SSIM 0.916 # 3
Video Super-Resolution UDM10 - 4x upscaling BasicVSR++ PSNR 40.72 # 3
SSIM 0.9722 # 3
Video Super-Resolution Vid4 - 4x upscaling BasicVSR++ PSNR 27.79 # 5
SSIM 0.8400 # 5
Video Super-Resolution Vid4 - 4x upscaling - BD degradation BasicVSR++ PSNR 29.04 # 3
SSIM 0.8753 # 3