Multi-Exposure Image Fusion: Perceptual Quality Assessment and Datasets

This paper, titled 'Perceptual Quality Assessment for Multi-Exposure Image Fusion' by Kede Ma, Kai Zeng, and Zhou Wang (published in IEEE Transactions on Image Processing, 24(11):3345–3356, 2015), explores the perceptual quality of multi-exposure image fusion techniques using two datasets:

1. Multi-Exposure Image Fusion Database: This dataset contains 30 pairs of multi-exposure images. Each pair consists of three images: a low-exposure image, a medium-exposure image, and a high-exposure image. These images are captured from real-world scenes.

2. Synthetic Dataset: This dataset also contains 30 pairs of multi-exposure images. Each pair consists of three images: a low-exposure image, a medium-exposure image, and a high-exposure image. These images are generated from real scenes using image synthesis techniques.