Recovering the shape of polyhedra using line-drawing analysis and complex reflectance models

被引:16
作者
Shimshoni, I
Ponce, J
机构
[1] UNIV ILLINOIS,DEPT COMP SCI,URBANA,IL 61801
[2] UNIV ILLINOIS,BECKMAN INST,URBANA,IL 61801
基金
美国国家科学基金会; 美国国家航空航天局;
关键词
D O I
10.1006/cviu.1996.0569
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Following K. Sugihara (Artif. Intell. 23, 1984, 59-95), we represent the geometric constraints imposed by the line-drawing of a polyhedron as a set of Linear equalities and inequalities. Unlike him, we explicitly take into account the uncertainty in vertex position, This allows us to circumvent the superstrictness of the constraints without deleting any of them, For a given error bound, deciding whether a line-drawing is the correct projection of a polyhedron is reduced to linear programming, and 3D shape recovery is reduced to optimization under linear constraints. Our method can be used for recovering the shape of any polyhedral object whose reflectance can be modelled accurately. We have implemented it for the following reflectance models: the Lambertian model, a Lambertian model with interreflections, and a reflectance model for specular objects, We present results obtained using real images. (C) 1997 Academic Press.
引用
收藏
页码:296 / 310
页数:15
相关论文
共 36 条
[1]  
Ballard D.H., 1982, Computer Vision
[2]  
Beckmann P., 1993, SCATTERING ELECTROMA
[3]   SEEING THINGS [J].
CLOWES, MB .
ARTIFICIAL INTELLIGENCE, 1971, 2 (01) :79-116
[4]  
Cohen M.F., 1985, P ACM SIGGRAPH 85, P31
[5]  
Cook R. L., 1981, Computer Graphics, V15, P307, DOI 10.1145/965161.806819
[6]   THE USE OF GRADIENT AND DUAL-SPACE IN LINE-DRAWING INTERPRETATION [J].
DRAPER, SW .
ARTIFICIAL INTELLIGENCE, 1981, 17 (1-3) :461-508
[7]   INTERPRETATION OF IMPERFECT LINE DATA AS A 3-DIMENSIONAL SCENE [J].
FALK, G .
ARTIFICIAL INTELLIGENCE, 1972, 3 (02) :101-144
[8]   REFLECTIONS ON SHADING [J].
FORSYTH, D ;
ZISSERMAN, A .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1991, 13 (07) :671-679
[9]  
GUZMAN A, 1968, MACTR59 MIT
[10]  
HEALEY G, 1987, P DARPA IM UND WORKS, P874