Skip to main content

Table 9 Main solutions arising from 2.5D reconstruction starting from painted images

From: Enhancing traditional museum fruition: current state and emerging tendencies

Cause

Effect using SFS methods

Solution

Possible drawbacks

Presence in the scene of a painted (i.e., guessed by the artist) diffuse illumination

Flattened final surface

Retrieval of a rough surface \({Z}_{rough}\) obtained as a result of inflating and smoothing iterative procedure. This allows to solve the SFS problem using only the principal illumination

Loss of details in the reconstructed surfaces. Too much emphasis on coarse volumes

Incorrect shading of objects due to artistic reproduction of subjects in the scene

Errors in shape reconstruction

Retrieval of a surface \({Z}_{main}\) based on SFS modified method using minimization and a set of boundary conditions robust to the possible errors in shading

If the shading of painted objects is grossly represented by the artist, the reconstruction may appear unfaithful

Incorrect scene principal illumination

Combined with the incorrect shading, this leads to errors in shape illumination

Use of an empirical method for determining, approximately, the illumination vector. Combining \({Z}_{main}\) with \({Z}_{rough}\) is equivalent to solve SFS problem using both diffuse illumination and principal illumination

Solving the method with more than one principal illumination leads to incorrect reconstructions

Loss of details due to excessive inflating and to over smoothing effect in SFS-based reconstruction (e.g., using high values of smoothness constraint)

None

Use of a refinement procedure allowing to consider finer details of the reconstructed objects

None