Generating Content for HDR Deghosting from Frequency View

TL;DR

This paper introduces LF-Diff, a compact diffusion model operating in a low-frequency latent space, significantly improving efficiency and detail in HDR image reconstruction from saturated and moving LDR images.

Contribution

The paper proposes a novel low-frequency aware diffusion model integrated with a regression-based HDR reconstruction network, enhancing efficiency and detail in ghost-free HDR imaging.

Findings

LF-Diff is 10 times faster than previous diffusion models.

The method outperforms state-of-the-art techniques on benchmark datasets.

Extensive experiments validate the effectiveness of the proposed approach.

Abstract

Recovering ghost-free High Dynamic Range (HDR) images from multiple Low Dynamic Range (LDR) images becomes challenging when the LDR images exhibit saturation and significant motion. Recent Diffusion Models (DMs) have been introduced in HDR imaging field, demonstrating promising performance, particularly in achieving visually perceptible results compared to previous DNN-based methods. However, DMs require extensive iterations with large models to estimate entire images, resulting in inefficiency that hinders their practical application. To address this challenge, we propose the Low-Frequency aware Diffusion (LF-Diff) model for ghost-free HDR imaging. The key idea of LF-Diff is implementing the DMs in a highly compacted latent space and integrating it into a regression-based model to enhance the details of reconstructed images. Specifically, as low-frequency information is closely related to human visual perception we propose to utilize DMs to create compact low-frequency priors for the reconstruction process. In addition, to take full advantage of the above low-frequency priors, the Dynamic HDR Reconstruction Network (DHRNet) is carried out in a regression-based manner to obtain final HDR images. Extensive experiments conducted on synthetic and real-world benchmark datasets demonstrate that our LF-Diff performs favorably against several state-of-the-art methods and is 10$\times$ faster than previous DM-based methods.