Synthetic faces, face cubes, and the geometry of face space

  title={Synthetic faces, face cubes, and the geometry of face space},
  author={Hugh R. Wilson and Gunter Loffler and Frances Wilkinson},
  journal={Vision Research},

Figures from this paper

A model of face space

Faces vary in their distinctiveness, and it has been suggested that—more distinctive faces are located further from the mean face in internal or perceptual face space (Valentine, 1991), than typical

Average faces: How does the averaging process change faces physically and perceptually

3D-face averaging, eye-movement tracking, and the computation of image-based face similarity set new constraints on the theoretical characterization of the average face and its role in establishing face norms but also offer practical guidance for creating approximated face norms to manipulate face identity.

Size-invariant but viewpoint-dependent representation of faces

Parametric face drawings: A demographically diverse and customizable face space model.

A novel face space model-parametric face drawings (or PFDs) are introduced to generate schematic, though realistic, parameterized line drawings of faces based on the statistical distribution of human facial features to generate morphs and caricatures.

The nature of synthetic face adaptation

Learning alters local face space geometry

Geometry of face space

It is shown here that a multidimensional Gaussian model of face images is a moderately successful approximation in describing symmetrized face subspace, and Random sampling of the Gaussianmodel for this subspace provides visually realistic synthetic faces, which respect gender.

Sub-exemplar shape tuning in human face-related areas.

The fusiform face area (FFA) revealed high face sensitivity, so that even facial images perceived as belonging to the same individual (<35%) were sufficient to produce full recovery from adaptation, and this results suggest that fMR-adaptation is closely related to behavior.

Contributions of individual face features to face discrimination




On The Other Side of the Mean: The Perception of Dissimilarity in Human Faces

The data indicate that there is a perceptual discontinuity of face identity as the face crosses over to the ‘other side of the mean’.

Low-dimensional representation of faces in higher dimensions of the face space

Various low-dimensional representations of the faces in the higher dimensions of the face space (i.e., the eigenvectors with smaller eigenvalues) provide better information for face recognition.

Face processing: Human perception and principal components analysis

It is concluded that shape and “texture” (the image-based information remaining after morphing) may be used separately by the human face processing system, and that PCA of images offers a useful tool for understanding this system.

Face recognition under varying poses: The role of texture and shape

Fine Discrimination of Faces can be Performed Rapidly

  • S. R. Lehky
  • Biology
    Journal of Cognitive Neuroscience
  • 2000
The behavioral evidence presented here indicates that complex visual processing can be completed within the first 100 msec of the signal, suggesting involvement of feedforward neural mechanisms, and placing constraints on possible computational algorithms employed within the ventral visual pathways.

Categorical perception of face identity in noise isolates configural processing.

The present study shows that configural processing can be isolated experimentally in normal subjects, and uses the CP-in-noise signature phenomenon to show thatconfigural processing survives image plane rotations of 45 degrees-90 degrees.

The Fusiform Face Area: A Module in Human Extrastriate Cortex Specialized for Face Perception

The data allow us to reject alternative accounts of the function of the fusiform face area (area “FF”) that appeal to visual attention, subordinate-level classification, or general processing of any animate or human forms, demonstrating that this region is selectively involved in the perception of faces.

Line Drawings of Faces Reduce Configural Processing

The highly specialised skill of face recognition found in humans is thought to be dependent on the processing of a combination of edge-based and surface-based information, and of single-feature as

A Unified Account of the Effects of Distinctiveness, Inversion, and Race in Face Recognition

  • T. Valentine
  • Psychology
    The Quarterly journal of experimental psychology. A, Human experimental psychology
  • 1991
It is argued that a multidimensional space provides a useful heuristic framework to investigate recognition of faces and that a norm-based coding version and the exemplar-based version of the framework cannot be distinguished on the basis of the experiments reported.

Eigenfaces for Recognition

A near-real-time computer system that can locate and track a subject's head, and then recognize the person by comparing characteristics of the face to those of known individuals, and that is easy to implement using a neural network architecture.