English

TransMEF: A Transformer-Based Multi-Exposure Image Fusion Framework using Self-Supervised Multi-Task Learning

Computer Vision and Pattern Recognition 2021-12-16 v3

Abstract

In this paper, we propose TransMEF, a transformer-based multi-exposure image fusion framework that uses self-supervised multi-task learning. The framework is based on an encoder-decoder network, which can be trained on large natural image datasets and does not require ground truth fusion images. We design three self-supervised reconstruction tasks according to the characteristics of multi-exposure images and conduct these tasks simultaneously using multi-task learning; through this process, the network can learn the characteristics of multi-exposure images and extract more generalized features. In addition, to compensate for the defect in establishing long-range dependencies in CNN-based architectures, we design an encoder that combines a CNN module with a transformer module. This combination enables the network to focus on both local and global information. We evaluated our method and compared it to 11 competitive traditional and deep learning-based methods on the latest released multi-exposure image fusion benchmark dataset, and our method achieved the best performance in both subjective and objective evaluations.

Keywords

Cite

@article{arxiv.2112.01030,
  title  = {TransMEF: A Transformer-Based Multi-Exposure Image Fusion Framework using Self-Supervised Multi-Task Learning},
  author = {Linhao Qu and Shaolei Liu and Manning Wang and Zhijian Song},
  journal= {arXiv preprint arXiv:2112.01030},
  year   = {2021}
}

Comments

Accepted by the Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI2022)

R2 v1 2026-06-24T08:01:01.275Z