Fig. 4
From: Unveiling the invisible: mathematical methods for restoring and interpreting illuminated manuscripts
Inpainting of damaged areas in Fig. 2. Once the inpainting domain is detected, the TV inpainting model (5, 6) is used to provide a good initialisation for the exemplar-based model (7). The final result shows the desired transfer of both geometric and texture information in the damaged areas. Patch size: \(5\times 5\) (upper row), \(7\times 7\) (bottom row)
