Fractal-IR: A Unified Framework for Efficient and Scalable Image Restoration
This addresses the problem of computational inefficiency and scalability in image restoration for researchers and practitioners, though it appears incremental as it builds on existing transformer methods with a novel architectural design.
The paper tackles the challenge of efficiently scaling vision transformers for multiple image restoration tasks by proposing Fractal-IR, a fractal-based framework that progressively refines images, achieving state-of-the-art performance across seven tasks with gains like 0.21 dB PSNR for super-resolution and 0.2 dB for denoising.
While vision transformers achieve significant breakthroughs in various image restoration (IR) tasks, it is still challenging to efficiently scale them across multiple types of degradations and resolutions. In this paper, we propose Fractal-IR, a fractal-based design that progressively refines degraded images by repeatedly expanding local information into broader regions. This fractal architecture naturally captures local details at early stages and seamlessly transitions toward global context in deeper fractal stages, removing the need for computationally heavy long-range self-attention mechanisms. Moveover, we observe the challenge in scaling up vision transformers for IR tasks. Through a series of analyses, we identify a holistic set of strategies to effectively guide model scaling. Extensive experimental results show that Fractal-IR achieves state-of-the-art performance in seven common image restoration tasks, including super-resolution, denoising, JPEG artifact removal, IR in adverse weather conditions, motion deblurring, defocus deblurring, and demosaicking. For $2\times$ SR on Manga109, Fractal-IR achieves a 0.21 dB PSNR gain. For grayscale image denoising on Urban100, Fractal-IR surpasses the previous method by 0.2 dB for $σ=50$.