HDR Imaging with Spatially Varying Signal-to-Noise Ratios

While today's high dynamic range (HDR) image fusion algorithms are capable of blending multiple exposures, the acquisition is often controlled so that the dynamic range within one exposure is narrow. For HDR imaging in photon-limited situations, the dynamic range can be enormous and the noise within one exposure is spatially varying. Existing image denoising algorithms and HDR fusion algorithms both fail to handle this situation, leading to severe limitations in low-light HDR imaging. This paper presents two contributions. Firstly, we identify the source of the problem. We find that the issue is associated with the co-existence of (1) spatially varying signal-to-noise ratio, especially the excessive noise due to very dark regions, and (2) a wide luminance range within each exposure. We show that while the issue can be handled by a bank of denoisers, the complexity is high. Secondly, we propose a new method called the spatially varying high dynamic range (SV-HDR) fusion network to simultaneously denoise and fuse images. We introduce a new exposure-shared block within our custom-designed multi-scale transformer framework. In a variety of testing conditions, the performance of the proposed SV-HDR is better than the existing methods.

1. Introduction

Today's high dynamic range (HDR) image fusion algorithms have demonstrated remarkable performances in blending images across a wide range of luminance levels. Many algorithms are able to handle an interior room with a sunlit view, of which the overall dynamic range is in the order of 100000:1 or more. However, most of these algorithms are designed for well-illuminated scenes. Even in the shortest exposure frame, the noise is maintained at a modest level so that the algorithm can focus on the blending task. The question we ask in this paper is: What if we push the shortest exposure to a photon-starving condition?