High Dynamic Range Imaging via Visual Attention Modules

TL;DR

This work tackles HDR reconstruction from LDR bursts by introducing a segmentation-guided Visual Attention Module (VAM). The model segments informative regions via Otsu thresholds, applies a six-stage deep architecture (feature extraction, VAM, spatial alignment, attention, reconstruction, refinement), and fuses information across exposures to produce high-quality HDR images. It demonstrates superior PSNR in HDR space and competitive tone-mapped metrics on a standard dataset, while acknowledging potential noise from segmentation and ghosting under motion. The approach offers a practical, segmentation-driven path to improve detail preservation in HDR while maintaining manageable computational demands.

Abstract

Thanks to High Dynamic Range (HDR) imaging methods, the scope of photography has seen profound changes recently. To be more specific, such methods try to reconstruct the lost luminosity of the real world caused by the limitation of regular cameras from the Low Dynamic Range (LDR) images. Additionally, although the State-Of-The-Art methods in this topic perform well, they mainly concentrate on combining different exposures and have less attention to extracting the informative parts of the images. Thus, this paper aims to introduce a new model capable of incorporating information from the most visible areas of each image extracted by a visual attention module (VAM), which is a result of a segmentation strategy. In particular, the model, based on a deep learning architecture, utilizes the extracted areas to produce the final HDR image. The results demonstrate that our method outperformed most of the State-Of-The-Art algorithms.

Figures (10)

• Figure 1: Produced masks of Short- and Long-Exposure images.
• Figure 2: The total pipeline of the proposed.
• Figure 3: The structure of the Feature Extraction Block.
• Figure 4: The structure of the Visual Attention Module (VAM).
• Figure 5: The structure of the Spatial Alignment Module.