Local Implicit Normalizing Flow for Arbitrary-Scale Image Super-Resolution
Flow-based methods have demonstrated promising results in addressing the ill-posed nature of super-resolution (SR) by learning the distribution of high-resolution (HR) images with the normalizing flow. However, these methods can only perform a predefined fixed-scale SR, limiting their potential in real-world applications. Meanwhile, arbitrary-scale SR has gained more attention and achieved great progress. Nonetheless, previous arbitrary-scale SR methods ignore the ill-posed problem and train the model with per-pixel L1 loss, leading to blurry SR outputs. In this work, we propose "Local Implicit Normalizing Flow" (LINF) as a unified solution to the above problems. LINF models the distribution of texture details under different scaling factors with normalizing flow. Thus, LINF can generate photo-realistic HR images with rich texture details in arbitrary scale factors. We evaluate LINF with extensive experiments and show that LINF achieves the state-of-the-art perceptual quality compared with prior arbitrary-scale SR methods.
PDF Abstract CVPR 2023 PDF CVPR 2023 AbstractCode
DIV2K
Urban100
Set14
Set5
Results from the Paper
Ranked #5 on
Image Super-Resolution
on DIV2K val - 4x upscaling
(using extra training data)
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Uses Extra Training Data |
Benchmark |
|---|---|---|---|---|---|---|---|
| Image Super-Resolution | DIV2K val - 4x upscaling | LINF | PSNR | 27.33 | # 11 | ||
| SSIM | 0.76 | # 13 | |||||
| LPIPS | 0.112 | # 5 | |||||
| Image Super-Resolution | DIV2K val - 4x upscaling | LINF t=0.0 | PSNR | 29.14 | # 3 | ||
| SSIM | 0.83 | # 4 | |||||
| LPIPS | 0.248 | # 9 |
