SGCM-Net: structure-guided mural inpainting with state space model
Yiyin Qiu, Jianjun Chen, Yan Fan, Md Rakib Hossain, Fei Wang, Xibei Yang · 2026
Read SGCM-Net: structure-guided mural inpainting with state space model by Yiyin Qiu, Jianjun Chen, Yan Fan, Md Rakib Hossain, Fei Wang, Xibei Yang (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/pmc-13408107.txt.
Description
Yiyin Qiu and five colleagues restore damaged ancient murals with a network that paints structure first and texture second, in a 2026 paper. Murals decay into tangled losses where both outlines and fine detail vanish, and standard networks see too narrow a window to stay globally consistent.
Their hybrid design fuses edge information from a first repair stage into every scale of the second. At its core, state space blocks scan the whole image with linear cost, capturing long-range relations that convolutions miss. On mural and landscape painting data sets, the method rebuilt coherent styles and details and matched leading approaches. The authors target heritage work where losses span large, complex areas.
Catalog record
| CREATOR | Yiyin Qiu, Jianjun Chen, Yan Fan, Md Rakib Hossain, Fei Wang, Xibei Yang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |