Simple Baselines for Image Restoration
This work addresses the need for simpler and more efficient image restoration models for researchers and practitioners, though it is incremental as it builds on existing architectures by removing nonlinear activations.
The paper tackles the problem of increasing complexity in state-of-the-art image restoration methods by proposing a simple baseline and a Nonlinear Activation Free Network (NAFNet), achieving SOTA results such as 33.69 dB PSNR on GoPro (exceeding previous SOTA by 0.38 dB with 8.4% computational cost) and 40.30 dB PSNR on SIDD (exceeding by 0.28 dB with less than half the cost).
Although there have been significant advances in the field of image restoration recently, the system complexity of the state-of-the-art (SOTA) methods is increasing as well, which may hinder the convenient analysis and comparison of methods. In this paper, we propose a simple baseline that exceeds the SOTA methods and is computationally efficient. To further simplify the baseline, we reveal that the nonlinear activation functions, e.g. Sigmoid, ReLU, GELU, Softmax, etc. are not necessary: they could be replaced by multiplication or removed. Thus, we derive a Nonlinear Activation Free Network, namely NAFNet, from the baseline. SOTA results are achieved on various challenging benchmarks, e.g. 33.69 dB PSNR on GoPro (for image deblurring), exceeding the previous SOTA 0.38 dB with only 8.4% of its computational costs; 40.30 dB PSNR on SIDD (for image denoising), exceeding the previous SOTA 0.28 dB with less than half of its computational costs. The code and the pre-trained models are released at https://github.com/megvii-research/NAFNet.