MEF-TBN: Multi-exposure image fusion approach based on a three-branch network for feature extraction and color enhancement

Chunmeng Wang, Wenxiang Zhang, Quan Zhang, Siyi Zhou · 2026

Read MEF-TBN: Multi-exposure image fusion approach based on a three-branch network for feature extraction and color enhancement by Chunmeng Wang, Wenxiang Zhang, Quan Zhang, Siyi Zhou (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0357156.txt.

Description

Chunmeng Wang and three colleagues fuse several photographs into one image that keeps both shadow and shine, in PLOS ONE in 2026. Scenes with bright skies and dark corners defeat a single exposure. Their system, MEF-TBN, splits the work among three cooperating branches. One reads fine local detail. One models the whole scene's structure with a transformer. The third restores natural color, which fusion methods often bleach or distort.

The branches produce weight maps that decide, pixel by pixel, which exposure should speak. A merging stage refines the local and global judgments together before the final blend. Against nine established fusion methods, the system topped most technical sharpness measures and swept the human perception oriented scores, while placing second on structural similarity. For phone cameras and medical imaging alike, the work shows that giving color its own dedicated branch pays off in pictures that look right, not just measure well.

Catalog record

CREATORChunmeng Wang, Wenxiang Zhang, Quan Zhang, Siyi Zhou
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69668
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0357156

Browse the full library · About the Tower of Babel