R. Epstein, P. Hallanan, A. L. Yuille, "52 Eigenimages Suffice: An Empirical Investigation of Low-Dimensional Lighting Models", IEEE Conf. Workshop on Physics-Based Vision, pp 108-116, 1995. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Face Recognition Under Variable Lighting using Harmonic.. - Lei Zhang Dimitris (2003) (3 citations) (Correct)
....subject instead. In general, appearance based methods like Eigenfaces [14] and SLAM [15] need a number of training images for each subject, in order to cope with illumination variability. Previous research suggests that illumination variability in face images is low dimensional e.g. 1] 2] 11][12][13] 20] Using spherical harmonics and signal processing techniques, Basri et al. 1] and Ramamoorthi [2] have shown that the set of images of a convex Lambertian object obtained under a wide variety of lighting conditions can be approximated accurately by a 9 dimensional linear subspace. ....
R.Epstein, P.Hallinan, and A.Yuille, "5+/-2 eigenimages suffice: An empirical investigation of low-dimensional lighting models," PBMCV, (SESSION 4), 1995
Efficient Region Tracking With Parametric Models of Geometry.. - Hager, al. (1998) (109 citations) (Correct)
....in space. This permits us to develop low order parametric models for the image motion of points within a target region models that can be used to predict the movement of the points and track the target through an image sequence. In the case of illumination, we exploit the observations of [17] [18], 19] to model image variation due to changing illumination by low dimensional linear subspaces. We then show that these models can be incorporated into an efficient estimation algorithm which establishes temporal correspondence of the target region by simultaneously determining both motion and ....
....case, complete loss of the target. Recently, it has been shown that a relatively small num1. Note that one possibility is that the gradient at a point is zero, in which case this is true of any motion. ber of basis images can often be used to account for large changes in illumination [19] [18], 17] 37] Briefly, the reason for this is as follows. Consider a point p on a Lambertian surface and a collimated light source characterized by a vector s R 3 , such that the direction of s gives the direction of the light rays and isi gives the intensity of the light source. The irradiance ....
R. Epstein, P. Hallinan, and A. Yuille, "5 2 Eigenimages Suffice: An Empirical Investigation of Low-Dimensional Lighting Models, " Technical Report 94-11, Harvard Univ., 1994.
Generalized Quotient Image - Haitao Wang Stan (2004) (Correct)
No context found.
R. Epstein, P. Hallanan, A. L. Yuille, "52 Eigenimages Suffice: An Empirical Investigation of Low-Dimensional Lighting Models", IEEE Conf. Workshop on Physics-Based Vision, pp 108-116, 1995.
Generalized Quotient Image - Haitao Wang Stan (2004) (Correct)
No context found.
R. Epstein, P. Hallanan, A. L. Yuille, "52 Eigenimages Suffice: An Empirical Investigation of Low-Dimensional Lighting Models", IEEE Conf. Workshop on Physics-Based Vision, pp 108-116, 1995.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE.. - Models Of Geometry (1997) (2 citations) (Correct)
No context found.
# R. Epstein, P. Hallinan, and A. Yuille, "5 2 Eigenimages Suffice: An Empirical Investigation of Low-Dimensional Lighting Models, " Technical Report 94-11, Harvard Univ., 1994.
Using Specularities for Recognition - Osadchy, Jacobs, Ramamoorthi (2003) (2 citations) (Correct)
No context found.
R. Epstein, P. Hallinan, A. Yuille. "5 2 eigenimages suffice: an empirical investigation of low-dimensional lighting models", IEEE Workshop on Physics-Based Vision: 108-116, 1995.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC