New constraints on data-closeness and needle map consistency for shape-from-shading

Citation
Pl. Worthington et Er. Hancock, New constraints on data-closeness and needle map consistency for shape-from-shading, IEEE PATT A, 21(12), 1999, pp. 1250-1267
Citations number
37
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN journal
01628828 → ACNP
Volume
21
Issue
12
Year of publication
1999
Pages
1250 - 1267
Database
ISI
SICI code
0162-8828(199912)21:12<1250:NCODAN>2.0.ZU;2-T
Abstract
This paper makes two contributions to the problem of needle-map recovery us ing shape-from-shading. First, we provide a geometric update procedure whic h allows the image irradiance equation to be satisfied as a hard constraint . This not only improves the data closeness of the recovered needle-map, bu t also removes the necessity for extensive parameter tuning. Second, we exp loit the improved ease of control of the new shape-from-shading process to investigate various types of needle-map consistency constraint. The first s et of constraints are based on needle-map smoothness. The second avenue of investigation is to use curvature information to impose topographic constra ints. Third, we explore ways in which the needle-map is recovered so as to be consistent with the image gradient field. In each case we explore a vari ety of robust error measures and consistency weighting schemes that can be used to impose the desired constraints on the recovered needle-map. We prov ide an experimental assessment of the new shape-from-shading framework on b oth real world images and synthetic images with known ground truth surface normals. The main conclusion drawn from our analysis is that the data-close ness constraint improves the efficiency of shape-from-shading and that both the topographic and gradient consistency constraints improve the fidelity of the recovered needle-map.