eduzhai > Applied Sciences > Engineering >

Benchmarking and Comparing Multi-exposure Image Fusion Algorithms

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 24 pages

Abstract: Multi-exposure image fusion (MEF) is an important area in computer vision andhas attracted increasing interests in recent years. Apart from conventionalalgorithms, deep learning techniques have also been applied to multi-exposureimage fusion. However, although much efforts have been made on developing MEFalgorithms, the lack of benchmark makes it difficult to perform fair andcomprehensive performance comparison among MEF algorithms, thus significantlyhindering the development of this field. In this paper, we fill this gap byproposing a benchmark for multi-exposure image fusion (MEFB) which consists ofa test set of 100 image pairs, a code library of 16 algorithms, 20 evaluationmetrics, 1600 fused images and a software toolkit. To the best of ourknowledge, this is the first benchmark in the field of multi-exposure imagefusion. Extensive experiments have been conducted using MEFB for comprehensiveperformance evaluation and for identifying effective algorithms. We expect thatMEFB will serve as an effective platform for researchers to compareperformances and investigate MEF algorithms.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×