Recovering Facial Shape Using a Statistical Model of Surface Normal Direction

Smith, William A. P.; Hancock, Edwin R.

doi:10.1109/tpami.2006.251

Cited by 101 publications

(57 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Azimuthal Equidistant Projection (AEP) is able to project normals onto points in a Euclidean space according to the direction. It has previously been employed to capture the local variations in facial shape [23], and can be applied to the normals, in order to project each 3D direction into the position in a Euclidean 2D plane. So for a regular grid of normals, defined as n(i, j) = (u i, j , v i, j , w i, j ), the AEP point p(i, j) = (x i, j , y i, j ) in this plane is defined as:…”

Section: Facial Geometry Representationsmentioning

confidence: 99%

Binary Pattern Analysis for 3D Facial Action Unit Detection

Sandbach

Zafeiriou

Pantić

2012

Procedings of the British Machine Vision Conference 2012

View full text Add to dashboard Cite

In this paper we propose new binary pattern features for use in the problem of 3D facial action unit (AU) detection. Two representations of 3D facial geometries are employed, the depth map and the Azimuthal Projection Distance Image (APDI). To these the traditional Local Binary Pattern is applied, along with Local Phase Quantisation, Gabor filters and Monogenic filters, followed by the binary pattern feature extraction method. Feature vectors are formed for each feature type through concatenation of histograms formed from the resulting binary numbers. Feature selection is then performed using a two-stage GentleBoost approach. Finally, we apply Support Vector Machines as classifiers for detection of each AU. This system is tested in two ways. First we perform 10-fold cross-validation on the Bosphorus database, and then we perform cross-database testing by training on this database and then testing on apex frames from the D3DFACS database, achieving promising results in both.

show abstract

Section: Facial Geometry Representationsmentioning

confidence: 99%

Binary Pattern Analysis for 3D Facial Action Unit Detection

Sandbach

Zafeiriou

Pantić

2012

Procedings of the British Machine Vision Conference 2012

View full text Add to dashboard Cite

show abstract

“…Subspace learning algorithms for normals, such as Principal Component Analysis (PCA), employ low-dimensional representation of surfaces [9,11,17,18]. In its simplest form, PCA on surface normals has been applied on the concatenation of normal coordinates [11].…”

Section: Introductionmentioning

confidence: 99%

“…In its simplest form, PCA on surface normals has been applied on the concatenation of normal coordinates [11]. One attempt to exploit the special structure of normals (i.e., that lie on a sphere) was conducted in [17]. In this paper the Azimuthal Equidistance Projection (AEP) was proposed and applied to surface normals prior to the application of PCA.…”

Section: Introductionmentioning

confidence: 99%

Robust Learning from Normals for 3D Face Recognition

Marras

Zafeiriou

Tzimiropoulos

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

Abstract. We introduce novel subspace-based methods for learning from the azimuth angle of surface normals for 3D face recognition. We show that the normal azimuth angles combined with Principal Component Analysis (PCA) using a cosine-based distance measure can be used for robust face recognition from facial surfaces. The proposed algorithms are well-suited for all types of 3D facial data including data produced by range cameras (depth images), photometric stereo (PS) and shade-from-X (SfX) algorithms. We demonstrate the robustness of the proposed algorithms both in 3D face reconstruction from synthetically occluded samples, as well as, in face recognition using the FRGC v2 3D face database and the recently collected Photoface database where the proposed method achieves state-of-the-art results. An important aspect of our method is that it can achieve good face recognition/verification performance by using raw 3D scans without any heavy preprocessing (i.e., model fitting, surface smoothing etc.).

show abstract

“…Facial shapes are synthesized from the most referenced patch pairs. Besides the example based approaches, more methods build statistical model on facial shapes [5,6,7,8,9,10] . Atick et al [5] first used statistical 3D face model to enhance SFS.…”

Section: Introductionmentioning

confidence: 99%

“…However human faces are not strictly symmetric objects, asymmetry of human faces results in errors in recovered shapes. Smith and Hancock [7] propose a statistic SFS approach that represents facial shape in surface normal domain. Surface normal directions are transformed into Cartesian points using azimuthal equidistant projection.…”

Section: Introductionmentioning

confidence: 99%

Recovering 3D facial shape via coupled 2D/3D space learning

Shan

Chen

et al. 2008

2008 8th IEEE International Conference on Automatic Face &Amp; Gesture Recognition

View full text Add to dashboard Cite

This paper presents a method for recovering 3D facial shape from single image via learning the relationship between the 2D intensity images and the 3D facial shapes. With a coupled training set, the intensity images and their corresponding facial shapes make up two vector spaces respectively. But only the correlated components in both spaces are useful for inference, so there must be embedded hidden subspaces in each space which preserve the interspace correlation information. Thus by learning the projection onto hidden subspaces based on Maximum Correlation Criteria and optimizing the linear transform between the hidden spaces, 3D facial shape is inferred from the intensity image. The effectiveness of the method is demonstrated on both synthesized and real world data.

show abstract

Recovering Facial Shape Using a Statistical Model of Surface Normal Direction

Abstract: Article:Smith, William A. P. orcid.org/0000-0002-6047-0413 and Hancock, Edwin R. orcid.org/0000-0003- 4496-2028 (2006) Recovering facial shape using a statistical model of surface normal direction. IEEE Transactions on Pattern Analysis and Machine

Cited by 101 publications

References 36 publications

Binary Pattern Analysis for 3D Facial Action Unit Detection

Binary Pattern Analysis for 3D Facial Action Unit Detection

Robust Learning from Normals for 3D Face Recognition

Recovering 3D facial shape via coupled 2D/3D space learning

Contact Info

Product

Resources

About