Improving 3D Facial Action Unit Detection with Intrinsic Normalization

Data normalization techniques are commonly used to reduce intra-class variance while preserving inter-class differences that aid in classification and regression tasks. Such practices bring all data examples to similar scale and range, and help to decrease the dimensionality of categorization problems. In this paper we describe a novel use of a geometric framework to extend the concept of data normalization to the domain of functions that lie on surfaces. In this context, normalization is posed as an embedding of all example functions into Riemannian manifolds nearly isometric to one another. Using geometric tools, we propose an implementation for the case of discretized functions on triangulated meshes. To demonstrate the proposed framework we apply it as a preprocessing step to the task of automatic facial action unit detection for depth maps from a 3D scanner. We show how this subject-independent representation produces more accurate action unit classification results. We also empirically demonstrate that utilizing this intrinsic normalization technique indeed improves action unit recognition performance.

[1]  P. Ekman Universal facial expressions of emotion. , 1970 .

[2]  P. Ekman,et al.  Facial action coding system: a technique for the measurement of facial movement , 1978 .

[3]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Kun Zhou,et al.  Mesh editing with poisson-based gradient field manipulation , 2004, SIGGRAPH 2004.

[5]  Markus Gross,et al.  Deformation Transfer for Detail-Preserving Surface Editing , 2006 .

[6]  Tieniu Tan,et al.  Combining Statistics of Geometrical and Correlative Features for 3D Face Recognition , 2006, BMVC.

[7]  Alexander M. Bronstein,et al.  Expression-Invariant Representations of Faces , 2007, IEEE Transactions on Image Processing.

[8]  Arman Savran,et al.  Bosphorus Database for 3D Face Analysis , 2008, BIOID.

[9]  Lijun Yin,et al.  Recognizing partial facial action units based on 3D dynamic range data for facial expression recognition , 2008, 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition.

[10]  Arman Savran,et al.  Automatic detection of facial actions from 3D data , 2009, 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops.

[11]  Luc Van Gool,et al.  Face/Off: live facial puppetry , 2009, SCA '09.

[12]  Emmanuel Dellandréa,et al.  AU recognition on 3D faces based on an extended statistical facial feature model , 2010, 2010 Fourth IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS).

[13]  Stefanos Zafeiriou,et al.  Binary Pattern Analysis for 3D Facial Action Unit Detection , 2012, BMVC.

[14]  Ioannis A. Kakadiaris,et al.  3D Facial Landmark Detection under Large Yaw and Expression Variations , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Matti Pietikäinen,et al.  CS-3DLBP and geometry based person independent 3D facial action unit detection , 2013, 2013 International Conference on Biometrics (ICB).