Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Perceptual Multi-Exposure Fusion

Xiaoning Liu

TL;DR

The paper tackles HDR imaging via multi-exposure fusion (MEF) with a focus on mobile-friendly efficiency. It replaces traditional saturation-based cues with Adaptive Well-Exposedness (AWE) and introduces a 3-D gradient for color images to preserve fine details, combining them within a Laplacian-pyramid fusion framework. A large static MEF benchmark (167 sequences) and a high-resolution photography (HRP) dataset are constructed to analyze parameter choices and guide design. Empirical results show the proposed perceptual MEF (PMEF) outperforms eight state-of-the-art methods in MEF-SSIM and visual detail preservation while maintaining lower computational complexity, underscoring its practicality for mobile HDR applications.

Abstract

As an ever-increasing demand for high dynamic range (HDR) scene shooting, multi-exposure image fusion (MEF) technology has abounded. In recent years, multi-scale exposure fusion approaches based on detail-enhancement have led the way for improvement in highlight and shadow details. Most of such methods, however, are too computationally expensive to be deployed on mobile devices. This paper presents a perceptual multi-exposure fusion method that not just ensures fine shadow/highlight details but with lower complexity than detailenhanced methods. We analyze the potential defects of three classical exposure measures in lieu of using detail-enhancement component and improve two of them, namely adaptive Wellexposedness (AWE) and the gradient of color images (3-D gradient). AWE designed in YCbCr color space considers the difference between varying exposure images. 3-D gradient is employed to extract fine details. We build a large-scale multiexposure benchmark dataset suitable for static scenes, which contains 167 image sequences all told. Experiments on the constructed dataset demonstrate that the proposed method exceeds existing eight state-of-the-art approaches in terms of visually and MEF-SSIM value. Moreover, our approach can achieve a better improvement for current image enhancement techniques, ensuring fine detail in bright light.

Perceptual Multi-Exposure Fusion

TL;DR

The paper tackles HDR imaging via multi-exposure fusion (MEF) with a focus on mobile-friendly efficiency. It replaces traditional saturation-based cues with Adaptive Well-Exposedness (AWE) and introduces a 3-D gradient for color images to preserve fine details, combining them within a Laplacian-pyramid fusion framework. A large static MEF benchmark (167 sequences) and a high-resolution photography (HRP) dataset are constructed to analyze parameter choices and guide design. Empirical results show the proposed perceptual MEF (PMEF) outperforms eight state-of-the-art methods in MEF-SSIM and visual detail preservation while maintaining lower computational complexity, underscoring its practicality for mobile HDR applications.

Abstract

As an ever-increasing demand for high dynamic range (HDR) scene shooting, multi-exposure image fusion (MEF) technology has abounded. In recent years, multi-scale exposure fusion approaches based on detail-enhancement have led the way for improvement in highlight and shadow details. Most of such methods, however, are too computationally expensive to be deployed on mobile devices. This paper presents a perceptual multi-exposure fusion method that not just ensures fine shadow/highlight details but with lower complexity than detailenhanced methods. We analyze the potential defects of three classical exposure measures in lieu of using detail-enhancement component and improve two of them, namely adaptive Wellexposedness (AWE) and the gradient of color images (3-D gradient). AWE designed in YCbCr color space considers the difference between varying exposure images. 3-D gradient is employed to extract fine details. We build a large-scale multiexposure benchmark dataset suitable for static scenes, which contains 167 image sequences all told. Experiments on the constructed dataset demonstrate that the proposed method exceeds existing eight state-of-the-art approaches in terms of visually and MEF-SSIM value. Moreover, our approach can achieve a better improvement for current image enhancement techniques, ensuring fine detail in bright light.
Paper Structure (30 sections, 26 equations, 14 figures, 5 tables)

This paper contains 30 sections, 26 equations, 14 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Multi-Scale Transform-Based Methods
  3. Statistical Model-Based and Patch-Based Methods
  4. Deep Learning-Based Methods
  5. Related Work
  6. Exposure Quality Metrics
  7. Contrast
  8. Saturation
  9. Well-exposedness
  10. Multi-Scale Fusion Framework
  11. Proposed Multi-exposure Image Fusion
  12. Rethinking Saturation
  13. Adaptive Well-Exposedness
  14. The Gradient of Color Images
  15. Multi-Scale Fusion Framework
  16. ...and 15 more sections

Figures (14)

  • Figure 1: Example results of several different MEF methods. (a) Mertens09 Mertens_2009. (b) S.Li13 Li_2013image. (c) Shen14 Shen_2014. (d) Kou17 kou2017multi. (e) Z.Li17 Li_2017. (f) Ma17 ma2017robust. (g) Wang19 wang2019detail. (h) Zhang19 zhang2020ifcnn. (i) Proposed. Green and red boxes are zoom-in regions for taking a closer look. One can see that our method retains more details in the highlights and shadow regions compared to Mertens09Mertens_2009.
  • Figure 2: The pipeline of the proposed MEF method.
  • Figure 3: A synthetic color image with a size of $256 \times 256$ where the pixel value of three channels in each row is the same and increases line by line from 0 to 255.
  • Figure 4: Some example images from our HRP dataset.
  • Figure 5: “Lamp” exposure sequence with different exposure time.
  • ...and 9 more figures