KBNet: Kernel Basis Network for Image Restoration
This work addresses adaptive spatial aggregation for image restoration, offering a more efficient alternative to transformers with improved performance, though it is incremental in combining kernel-based and attention mechanisms.
The paper tackled the problem of adaptive spatial information aggregation in image restoration by proposing a kernel basis network (KBNet), which uses learnable kernel bases and multi-axis feature fusion to achieve state-of-the-art performance on over ten benchmarks for tasks like denoising, deraining, and deblurring while reducing computational costs.
How to aggregate spatial information plays an essential role in learning-based image restoration. Most existing CNN-based networks adopt static convolutional kernels to encode spatial information, which cannot aggregate spatial information adaptively. Recent transformer-based architectures achieve adaptive spatial aggregation. But they lack desirable inductive biases of convolutions and require heavy computational costs. In this paper, we propose a kernel basis attention (KBA) module, which introduces learnable kernel bases to model representative image patterns for spatial information aggregation. Different kernel bases are trained to model different local structures. At each spatial location, they are linearly and adaptively fused by predicted pixel-wise coefficients to obtain aggregation weights. Based on the KBA module, we further design a multi-axis feature fusion (MFF) block to encode and fuse channel-wise, spatial-invariant, and pixel-adaptive features for image restoration. Our model, named kernel basis network (KBNet), achieves state-of-the-art performances on more than ten benchmarks over image denoising, deraining, and deblurring tasks while requiring less computational cost than previous SOTA methods.