HDR Image Reconstruction using an Unsupervised Fusion Model

TL;DR

This paper introduces an unsupervised deep learning method for HDR image reconstruction from multi-exposure LDR images, effectively combining details from different exposures without needing ground-truth HDR data.

Contribution

It presents a novel unsupervised CNN-based fusion model for HDR imaging that outperforms existing methods and does not require HDR ground-truth images for training.

Findings

Achieves higher MEF-SSIM scores than existing methods.

Effectively combines details from underexposed and overexposed images.

Demonstrates practical applicability without ground-truth HDR data.

Abstract

High Dynamic Range (HDR) imaging aims to reproduce the wide range of brightness levels present in natural scenes, which the human visual system can perceive but conventional digital cameras often fail to capture due to their limited dynamic range. To address this limitation, we propose a deep learning-based multi-exposure fusion approach for HDR image generation. The method takes a set of differently exposed Low Dynamic Range (LDR) images, typically an underexposed and an overexposed image, and learns to fuse their complementary information using a convolutional neural network (CNN). The underexposed image preserves details in bright regions, while the overexposed image retains information in dark regions; the network effectively combines these to reconstruct a high-quality HDR output. The model is trained in an unsupervised manner, without relying on ground-truth HDR images, making it practical for real-world applications where such data is unavailable. We evaluate our results using the Multi-Exposure Fusion Structural Similarity Index Measure (MEF-SSIM) and demonstrate that our approach achieves superior visual quality compared to existing fusion methods. A customized loss function is further introduced to improve reconstruction fidelity and optimize model performance.