Search papers, labs, and topics across Lattice.
This paper introduces Hybrid-LUT, a novel YUV-based framework for image denoising that integrates a multi-band lookup table (LUT) for the luminance channel with lightweight filtering for chrominance channels. By reducing LUT storage by two-thirds compared to traditional RGB-LUT methods, Hybrid-LUT achieves state-of-the-art performance while maintaining runtime efficiency. Extensive evaluations reveal that it surpasses existing LUT-based approaches by at least 0.63 dB CPSNR on real-world datasets, making it particularly suitable for resource-constrained edge devices.
Hybrid-LUT slashes LUT storage by two-thirds while enhancing image denoising quality, outperforming existing methods by a significant margin.
Lookup table (LUT)-based image denoising methods have attracted increasing attention due to their high efficiency and hardware-friendly properties. However, existing RGB-LUT approaches require three identical LUTs to process RGB channels in parallel, resulting in large on-chip SRAM consumption. A simple alternative is to apply LUT processing only to the luminance (Y) channel in the YUV color space to reduce memory usage. However, this naive strategy leads to degraded restoration quality, since ignoring the chrominance (UV) channels introduces color distortion and residual artifacts. In this work, we propose Hybrid-LUT, a YUV-based asymmetric channel-processing framework that combines LUT and filtering in a unified design. Specifically, a multi-band LUT branch with pixel-level weight fusion is applied to the Y channel to recover fine textures, while lightweight filtering is used for the UV channels to maintain color consistency. This design reduces LUT storage by two-thirds compared with RGB-LUT methods while maintaining the same runtime throughput. Extensive experiments show that Hybrid-LUT achieves state-of-the-art (SOTA) performance across multiple benchmarks with only 421 KB of storage. In particular, our method surpasses existing LUT-based denoising approaches by at least 0.63 dB CPSNR on real-world datasets, demonstrating its effectiveness for image denoising on resource-constrained edge devices. The project is available at https://github.com/Ai-ZL/Hybrid-LUT .
