Lambertian reflectance and linear subspaces

Basri, Ronen; Jacobs, David W.

doi:10.1109/tpami.2003.1177153

Cited by 1,492 publications

(924 citation statements)

References 29 publications

Supporting

Mentioning

915

Contrasting

Unclassified

Order By: Relevance

“…For a fully automatic parameter choice, a quality metric images should be used: σ Ph and σ H should be chosen in such a way that not two much information of image is removed. 2 http://uni-lj.academia.edu/VitomirStruc 3 The code for this method is available from http://parnec.nuaa.edu.cn/xtan/ (Wang et al, 2004) 100 100 98.3 88.5 79.5 PS (Tan & Triggs, 2007) 100 100 98.4 97.9 96.7 LTV (Chen et al, 2006) + 100 100 100 100 100 Retina filter 100 100 100 100 100 Cone-cast (Georghiades & Belhumeur, 2001)* 100 100 100 100 -Harmonic image (Basri & Jacobs, 2003 Table 2 that:…”

Section: Results On the Extended Yale B Databasementioning

confidence: 99%

“…A training phase is then performed so as to derive a model for every identy, which will be used for recognition task. Examples are Illumination Cone (Belhumeur & Kriegman, 1998), Spherical Harmonics (Basri & Jacobs, 2003). Although providing the high quality results in general, these algorithms are costly and in particular they require several images obtained under different lighting conditions for each individual to be recognized.…”

Section: Illumination Modelingmentioning

confidence: 99%

See 1 more Smart Citation

Biologically Inspired Processing for Lighting Robust Face Recognition

Vu¹,

Caplier²

2011

State of the Art in Biometrics

View full text Add to dashboard Cite

Section: Results On the Extended Yale B Databasementioning

confidence: 99%

Section: Illumination Modelingmentioning

confidence: 99%

Biologically Inspired Processing for Lighting Robust Face Recognition

Vu¹,

Caplier²

2011

State of the Art in Biometrics

View full text Add to dashboard Cite

“…As demonstrated in [2], the illumination space of a Lambertian object is well approximated by a low-dimensional linear space. This implies that if D represents a data set consisting of digital images of a fixed Lambertian object collected under a variety of illumination conditions and with a fixed resolution, then a very high percentage of the energy of D is captured by a low-dimensional linear space inside the vector space generated by all possible digital images at the same fixed resolution.…”

Section: Compression Of N Inmentioning

confidence: 99%

“…Since illumination spaces can be well-approximated by a 10-dimensionallinear subspaces [2,38], we randomly select two disjoint sets of size ten for the points in the probe and gallery. This process is repeated ten times producing a total of 670 probe points.…”

Section: Compressions Of Nmentioning

confidence: 99%

See 1 more Smart Citation

Feature Patch Illumination Spaces and Karcher Compression for Face Recognition via Grassmannians

Chang¹,

Peterson²,

Kirby³

2012

APM

View full text Add to dashboard Cite

Recent work has established that digital images of a human face, when collected with a fixed pose but under a variety of illumination conditions, possess discriminatory information that can be used in classification. In this paper we perform classification on Grassmannians to demonstrate that sufficient discriminatory information persists in feature patch (e.g., nose or eye patch) illumination spaces. We further employ the use of Karcher mean on the Grassmannians to demonstrate that this compressed representation can accelerate computations with relatively minor sacrifice on performance. The combination of these two ideas introduces a novel perspective in performing face recognition.

show abstract

Combining Geometrical and Statistical Models for Video-Based Face Recognition

View full text Add to dashboard Cite

A number of methods in tracking and recognition have successfully exploited lowdimensional representations of object appearance learned from a set of examples. In all these approaches, the construction of the underlying low-dimensional manifold relied upon obtaining different instances of the object's appearance and then using statistical data analysis tools to approximate the appearance space. This requires collecting a very large number of examples and the accuracy of the method depends upon the examples that have been chosen. In this chapter, we show that it is possible to estimate low-dimensional manifolds that describe object appearance using a combination of analytically derived geometrical models and statistical data analysis. Specifically, we derive a quadrilinear space of object appearance that is able to represent the effects of illumination, motion, identity and shape. We then show how efficient tracking algorithms like inverse compositional estimation can be adapted to the geometry of this manifold. Our proposed method significantly reduces the amount of data that needs to be collected for learning the manifolds and makes the learned manifold less dependent upon the actual examples that were used. Based upon this novel manifold, we present a framework for face recognition from video sequences that is robust to large changes in facial pose and lighting conditions. The method can handle situations where the pose and lighting conditions in the training and testing data are completely disjoint. We show detailed performance analysis results and recognition scores on a large video dataset.

show abstract

Lambertian reflectance and linear subspaces

Cited by 1,492 publications

References 29 publications

Biologically Inspired Processing for Lighting Robust Face Recognition

Biologically Inspired Processing for Lighting Robust Face Recognition

Feature Patch Illumination Spaces and Karcher Compression for Face Recognition via Grassmannians

Combining Geometrical and Statistical Models for Video-Based Face Recognition

Contact Info

Product

Resources

About