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, Khan Bahadar Khan · 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, Khan Bahadar Khan (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/PMC13592636.txt.
Description
Chunmeng Wang and four colleagues build a network that fuses several exposures of one scene into a single wide range image. Older fusion methods tend to lose fine detail, global structure, or true color. Their design runs three branches in parallel: one reads local features with attention, one reads global links with a transformer, and one restores natural color. Merging blocks blend the first two branches' weight maps before guided filtering lifts them to full size.
Against nine rival methods on three datasets, the network topped most detail and human perception scores and placed second on a structural similarity measure. Gains over a leading transformer method reached about twenty percent on one visual quality score. Wang and the team present the three branch split as the fix for detail and color losses.
Catalog record
| CREATOR | Chunmeng Wang, Wenxiang Zhang, Quan Zhang, Siyi Zhou, Khan Bahadar Khan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |