Adjustable Complexity Transformer Architecture for Image Denoising
Open Access
Abstract
In recent years, image denoising has seen a shift from traditional non-local self-similarity methods like BM3D to deep-learning based approaches that use learnable convolutions and attention mechanisms. While pixel-level attention is effective at capturing long-range relationships similar to non-local self-similarity based methods, it incurs extremely high computational costs that scale quadratically with image resolution. As an alternative, channel-wise attention is resolution-independent and computationally efficient but may miss crucial spatial details. In this paper, an adjustable attention mechanism is introduced that bridges the gap between pixel and channel attentions. In the proposed model, average pooling and variable-size convolutions are added before attention calculation to adjust spatial resolution and, thus, allow dynamical adjustment of computational complexity. This adjustable attention is applied in a transformer-based U-Net architecture and achieves performance comparable to state-of-the-art methods in both real and Gaussian blind denoising tasks. To be more concrete, the proposed method achieves a Peak Signal-to-Noise Ratio of 39.65 dB and a Structural Similarity Index Measure of 0.913 on the Smartphone Image Denoising Dataset. Therefore, the proposed method demonstrates a balance between efficiency and denoising quality.Keywords
This publication has 64 references indexed in Scilit:
- Image blind denoising using dual convolutional neural network with skip connectionSignal Processing: Image Communication, 2025
- Eficient image denoising using deep learning: A brief surveyInformation Fusion, 2025
- When Fast Fourier Transform Meets Transformer for Image Restoration2024
- Dual convolutional neural network with attention for image blind denoisingMultimedia Systems, 2024
- EWT: Efficient Wavelet-Transformer for single image denoisingNeural Networks, 2024
- Pyramid Attention Network for Image RestorationInternational Journal of Computer Vision, 2023
- Simple Baselines for Image Restoration2022
- Improving non-local means image denoising by correlation correctionMultidimensional Systems and Signal Processing, 2022
- Attention-guided CNN for image denoisingNeural Networks, 2020
- Image denoising using deep CNN with batch renormalizationNeural Networks, 2020