A Comparative Study of NAFNet Baselines for Image Restoration
This is an incremental study for researchers in image restoration, providing empirical validation of NAFNet's design choices.
The study tackled image restoration by evaluating NAFNet's core components on corrupted CIFAR10 images, finding that SimpleGate activation and simplified attention mechanisms yield better PSNR and SSIM results than conventional methods.
We study NAFNet (Nonlinear Activation Free Network), a simple and efficient deep learning baseline for image restoration. By using CIFAR10 images corrupted with noise and blur, we conduct an ablation study of NAFNet's core components. Our baseline model implements SimpleGate activation, Simplified Channel Activation (SCA), and LayerNormalization. We compare this baseline to different variants that replace or remove components. Quantitative results (PSNR, SSIM) and examples illustrate how each modification affects restoration performance. Our findings support the NAFNet design: the SimpleGate and simplified attention mechanisms yield better results than conventional activations and attention, while LayerNorm proves to be important for stable training. We conclude with recommendations for model design, discuss potential improvements, and future work.