Efficient Burst Super-Resolution with One-step Diffusion

Kento Kawai, Takeru Oba, Kyotaro Tokoro, Kazutoshi Akita, Norimichi Ukita; Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR) Workshops, 2025, pp. 864-873

Abstract

While burst Low-Resolution (LR) images are useful for improving their Super Resolution (SR) image compared to a single LR image, prior burst SR methods are trained in a deterministic manner, which produces a blurry SR image. Since such blurry images are perceptually degraded, we aim to reconstruct sharp and high-fidelity SR images by a diffusion model. Our method improves the efficiency of the diffusion model with a stochastic sampler with a high-order ODE as well as one-step diffusion using knowledge distillation. Our experimental results demonstrate that our method can reduce the runtime to 1.6 % of its baseline while maintaining the SR quality measured based on image distortion and perceptual quality.

Related Material

[pdf]
[bibtex]
@InProceedings{Kawai_2025_CVPR, author = {Kawai, Kento and Oba, Takeru and Tokoro, Kyotaro and Akita, Kazutoshi and Ukita, Norimichi}, title = {Efficient Burst Super-Resolution with One-step Diffusion}, booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR) Workshops}, month = {June}, year = {2025}, pages = {864-873} }