G. Carlomagno, F. Renna, N. Mosca, G. Attolico, and A. Distante (Italy)
Fragmented frescos, Virtual aided recomposition, CBIR, Color correction, Color pairs extraction.
The paper describes the color correction approach used in a system for the virtual aided recomposition of fragmented frescos, being developed and proved for the reconstruction of the St. Matthew's fresco, painted by Cimabue for the Upper Church of St. Francis in Assisi and broken in more than 140.000 fragments during the earthquake of 1997. One of its key features is the selection of fragments using a query-by-example approach applied to suitable sample images (other fragments and/or details extracted from the whole image of the fresco). This approach is especially challenging because the only available picture of the whole fresco has been acquired without any care about color calibration: its comparison with fragments requires the application of an appropriate color transformation to each part of the whole image. Two methods to estimate this correction, one requiring the manual extraction of color pairs and the other working automatically on the already placed fragments, have been developed and compared. Experimental results show that the automatic approach reaches satisfactory results at a much lower level of complexity. The obtained improvement in terms of color similarity results is shown.
Important Links:
Go Back
