Results 1  10
of
474,343
TheBasReliefAmbiguity PeterN.Belhumeur DavidJ.Kriegmany AlanL.Yuille
"... Sinceantiquity,artisanshavecreatedattenedforms, oftencalled\basreliefs,"whichgiveanexaggerated perceptionofdepthwhenviewedfromaparticularvantagepoint.Thispaperpresentsanexplanationof thisphenomena,showingthattheambiguityindeterminingthereliefofanobjectisnotconnedto basreliefsculpturebutisimpl ..."
Abstract
 Add to MetaCart
Sinceantiquity,artisanshavecreatedattenedforms, oftencalled\basreliefs,"whichgiveanexaggerated perceptionofdepthwhenviewedfromaparticularvantagepoint.Thispaperpresentsanexplanationof thisphenomena,showingthattheambiguityindeterminingthereliefofanobjectisnotconnedto basreliefsculpturebutisimplicitinthedeterminationofthestructureofanyobject.Formally,ifthe object'struesurfaceisdenotedbyztrue=f(x;y), thenwedenethe\generalizedbasrelieftransformation"asz=f(x;y)+x+y;withacorresponding transformationofthealbedo.Foreachimageofa Lambertiansurfacef(x;y)producedbyapointlight sourceatinnity,thereexistsanidenticalimageofa basreliefproducedbyatransformedlightsource.This equalityholdsforbothshadedandshadowedregions. Thus,thesetofpossibleimages(illuminationcone)is invariantovergeneralizedbasrelieftransformations.
KriegmanBelhumeur Vision Technologies
"... We present a novel approach to localizing parts in images of human faces. The approach combines the output of local detectors with a nonparametric set of global models for the part locations based on over one thousand handlabeled exemplar images. By assuming that the global models generate the par ..."
Abstract
 Add to MetaCart
We present a novel approach to localizing parts in images of human faces. The approach combines the output of local detectors with a nonparametric set of global models for the part locations based on over one thousand handlabeled exemplar images. By assuming that the global models generate the part locations as hidden variables, we derive a Bayesian objective function. This function is optimized using a consensus of models for these hidden variables. The resulting localizer handles a much wider range of expression, pose, lighting and occlusion than prior ones. We show excellent performance on a new dataset gathered from the internet and show that our localizer achieves stateoftheart performance on the less challenging BioID dataset. 1.
Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection
, 1997
"... We develop a face recognition algorithm which is insensitive to gross variation in lighting direction and facial expression. Taking a pattern classification approach, we consider each pixel in an image as a coordinate in a highdimensional space. We take advantage of the observation that the images ..."
Abstract

Cited by 2263 (18 self)
 Add to MetaCart
We develop a face recognition algorithm which is insensitive to gross variation in lighting direction and facial expression. Taking a pattern classification approach, we consider each pixel in an image as a coordinate in a highdimensional space. We take advantage of the observation that the images of a particular face, under varying illumination but fixed pose, lie in a 3D linear subspace of the high dimensional image space  if the face is a Lambertian surface without shadowing. However, since faces are not truly Lambertian surfaces and do indeed produce selfshadowing, images will deviate from this linear subspace. Rather than explicitly modeling this deviation, we linearly project the image into a subspace in a manner which discounts those regions of the face with large deviation. Our projection method is based on Fisher's Linear Discriminant and produces well separated classes in a lowdimensional subspace even under severe variation in lighting and facial expressions. The Eigenface
Dropout from higher education: A theoretical synthesis of recent research
 Review of Educational Research
, 1975
"... Despite the very extensive literature on dropout from higher education, much remains unknown about the nature of the dropout process. In large measure, the failure of past research to delineate more clearly the multiple characteristics of dropout can be traced to two major shortcomings; namely, inad ..."
Abstract

Cited by 755 (2 self)
 Add to MetaCart
. Nor is it uncommon to find permanent dropouts placed together with persons whose leaving may be temporary in I am indebted to my research assistant, John B. Cullen, for having made an extensive literature search and compiling summaries of the literature for me. I am also indebted to Professors Peter
A Fast Quantum Mechanical Algorithm for Database Search
 ANNUAL ACM SYMPOSIUM ON THEORY OF COMPUTING
, 1996
"... Imagine a phone directory containing N names arranged in completely random order. In order to find someone's phone number with a probability of , any classical algorithm (whether deterministic or probabilistic)
will need to look at a minimum of names. Quantum mechanical systems can be in a supe ..."
Abstract

Cited by 1126 (10 self)
 Add to MetaCart
Imagine a phone directory containing N names arranged in completely random order. In order to find someone's phone number with a probability of , any classical algorithm (whether deterministic or probabilistic)
will need to look at a minimum of names. Quantum mechanical systems can be in a
Lambertian Reflectance and Linear Subspaces
, 2000
"... We prove that the set of all reflectance functions (the mapping from surface normals to intensities) produced by Lambertian objects under distant, isotropic lighting lies close to a 9D linear subspace. This implies that, in general, the set of images of a convex Lambertian object obtained under a wi ..."
Abstract

Cited by 514 (20 self)
 Add to MetaCart
We prove that the set of all reflectance functions (the mapping from surface normals to intensities) produced by Lambertian objects under distant, isotropic lighting lies close to a 9D linear subspace. This implies that, in general, the set of images of a convex Lambertian object obtained under a wide variety of lighting conditions can be approximated accurately by a lowdimensional linear subspace, explaining prior empirical results. We also provide a simple analytic characterization of this linear space. We obtain these results by representing lighting using spherical harmonics and describing the effects of Lambertian materials as the analog of a convolution. These results allow us to construct algorithms for object recognition based on linear methods as well as algorithms that use convex optimization to enforce nonnegative lighting functions. Finally, we show a simple way to enforce nonnegative lighting when the images of an object lie near a 4D linear space. Research conducted w...
An introduction to Kolmogorov Complexity and its Applications: Preface to the First Edition
, 1997
"... This document has been prepared using the L a T E X system. We thank Donald Knuth for T E X, Leslie Lamport for L a T E X, and Jan van der Steen at CWI for online help. Some figures were prepared by John Tromp using the xpic program. The London Mathematical Society kindly gave permission to reproduc ..."
Abstract

Cited by 2143 (120 self)
 Add to MetaCart
to reproduce a long extract by A.M. Turing. The Indian Statistical Institute, through the editor of Sankhy¯a, kindly gave permission to quote A.N. Kolmogorov. We gratefully acknowledge the financial support by NSF Grant DCR8606366, ONR Grant N0001485k0445, ARO Grant DAAL0386K0171, the Natural Sciences
A New Method for Solving Hard Satisfiability Problems
 AAAI
, 1992
"... We introduce a greedy local search procedure called GSAT for solving propositional satisfiability problems. Our experiments show that this procedure can be used to solve hard, randomly generated problems that are an order of magnitude larger than those that can be handled by more traditional approac ..."
Abstract

Cited by 734 (21 self)
 Add to MetaCart
approaches such as the DavisPutnam procedure or resolution. We also show that GSAT can solve structured satisfiability problems quickly. In particular, we solve encodings of graph coloring problems, Nqueens, and Boolean induction. General application strategies and limitations of the approach are also
Where the REALLY Hard Problems Are
 IN J. MYLOPOULOS AND R. REITER (EDS.), PROCEEDINGS OF 12TH INTERNATIONAL JOINT CONFERENCE ON AI (IJCAI91),VOLUME 1
, 1991
"... It is well known that for many NPcomplete problems, such as KSat, etc., typical cases are easy to solve; so that computationally hard cases must be rare (assuming P != NP). This paper shows that NPcomplete problems can be summarized by at least one "order parameter", and that the hard p ..."
Abstract

Cited by 681 (1 self)
 Add to MetaCart
problems. We show that for some P problems either there is no phase transition or it occurs for bounded N (and so bound...
Social capital, intellectual capital, and the organizational advantage
 Academy of Management Review
, 1998
"... Scholars of the theory of the firm have begun to emphasize the sources and conditions of what has been described a s "the organizational advantage, " rather than focus on the causes and consequences of market failure. Typically, researchers see such organizational advantage a s accruing fr ..."
Abstract

Cited by 1100 (1 self)
 Add to MetaCart
settings. a re conducive to the development of high levels of social capital. and (3) it is because of their more dense social capital that firms. within certain limits. have a n advantage over markets in creating and sharing intellectual capital. We present a model that incorporates this overall argument
Results 1  10
of
474,343