Towards Real-World Burst Image Super-Resolution: Benchmark and Method
This work addresses the challenge of burst image super-resolution for real-world applications, providing a dataset and method that improve reconstruction quality, though it appears incremental as it builds on existing burst SR techniques.
The authors tackled the problem of reconstructing high-quality images from multiple low-resolution frames in real-world scenarios by introducing a new dataset (RealBSR) and a method (FBAnet) that outperforms existing state-of-the-art burst super-resolution methods, achieving visually-pleasant results with model details.
Despite substantial advances, single-image super-resolution (SISR) is always in a dilemma to reconstruct high-quality images with limited information from one input image, especially in realistic scenarios. In this paper, we establish a large-scale real-world burst super-resolution dataset, i.e., RealBSR, to explore the faithful reconstruction of image details from multiple frames. Furthermore, we introduce a Federated Burst Affinity network (FBAnet) to investigate non-trivial pixel-wise displacements among images under real-world image degradation. Specifically, rather than using pixel-wise alignment, our FBAnet employs a simple homography alignment from a structural geometry aspect and a Federated Affinity Fusion (FAF) strategy to aggregate the complementary information among frames. Those fused informative representations are fed to a Transformer-based module of burst representation decoding. Besides, we have conducted extensive experiments on two versions of our datasets, i.e., RealBSR-RAW and RealBSR-RGB. Experimental results demonstrate that our FBAnet outperforms existing state-of-the-art burst SR methods and also achieves visually-pleasant SR image predictions with model details. Our dataset, codes, and models are publicly available at https://github.com/yjsunnn/FBANet.