SGDFormer: One-stage transformer-based architecture for cross-spectral stereo image guided denoising

Runmin Zhang, Zhu Yu, Zehua Sheng, Jiacheng Ying, Si-Yuan Cao, Shu-Jie Chen, Bailin Yang, Junwei Li, Hui-Liang Shen

Abstract: Highlights•We propose a novel one-stage transformer-based architecture for cross-spectral stereo image guided denoising.•We design a noise-robust cross-attention module to capture precise stereo correspondence.•We devise a simple but effective spatially variant feature fusion strategy.•Our method achieves state-of-the-art denoising performance on various datasets.