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

@article{Troje1996FaceRU,
  title={Face recognition under varying poses: The role of texture and shape},
  author={Nikolaus F. Troje and Heinrich H. B{\"u}lthoff},
  journal={Vision Research},
  year={1996},
  volume={36},
  pages={1761-1771}
}
Although remarkably robust, face recognition is not perfectly invariant to pose and viewpoint changes. It has long been known that both profile and full-face views result in poorer recognition performance than a 3/4 view. However, little data exist which investigate this phenomenon in detail. The present work provides such data using a high angular resolution and a large range of poses. Since there are inconsistencies in the literature concerning these issues, we emphasize the different roles… 
Is face recognition in pictures affected by the center of projection?
  • C. Liu
  • Computer Science
    2003 IEEE International SOI Conference. Proceedings (Cat. No.03CH37443)
  • 2003
TLDR
The results showed little support for the prediction that the same position should help correct marginal distortions of shapes due to large perspective convergence and hence facilitate reconstruction of 3D shape from perspective cues, and the lack of3D shape reconstruction and the effects of image similarity provided favorable evidence for the image-based theory of face recognition.
Efficient visual information for unfamiliar face matching despite viewpoint variations: It’s not in the eyes!
TLDR
The results indicate that the use of features located in thecenter of the face, the lower left portion of the nose area and the center of the mouth, are significantly associated with individual efficiency to generalize a face's identity across different viewpoints, and suggests that a certain distinction can be made between the visual mechanisms subtending viewpoint invariance and face recognition in the case of unfamiliar face identification.
Feature processing and superiority of three-quarter views in face recognition
TLDR
A positive effect of the makeup cue is found, which could be due to better recognition when the shape of eyes and lips are highlighted, and better recognition with left- sided views compared to right-sided views.
Multiple Representations in the Recognition of Faces at Novel Orientations
Bilaterally-symmetric objects such as faces are easily recognized from an orientation that is symmetric with respect to the original view. Several authors have proposed that a simple image-based
Information and viewpoint dependence in face recognition
TLDR
Three experiments used three-dimensional facial surface representations to investigate the effects of rotations in depth on a face recognition task to show that three- dimensional shape information is fundamental for recognition across rotation in depth although superficial information may also be used to reduce viewpoint dependence.
Using 3D computer graphics for perception: the role of local and global information in face processing
TLDR
The results demonstrated that human observers are capable of recognizing faces across different viewpoints on the sole basis of isolated featural information and of isolated configural information.
Stimulus-specific effects in face recognition over changes in viewpoint
TLDR
The results suggest that the stimulus information captured in the model measures may underlie distinct and dissociable aspects of the recognizability of individual faces across viewpoint change.
The effect of texture on face identification and configural information processing
Shape and texture are an integral part of face identity. In the present study, the importance of face texture for face identification and detection of configural manipulation (i.e., spatial relation
Are face representations viewpoint dependent? A stereo advantage for generalising across different views of faces
TLDR
The current experiment demonstrates that viewpoint costs are reduced by depicting the face with stereoscopic three-dimensionality (compared to a synoptically presented face), raising the possibility that the viewpoint costs found in face recognition experiments might be a better reflection of the information that is typically unavailable in the experimental stimuli than of the underlying neural representation of facial identity.
Advantages of 3D Methods for Face Recognition Research in Humans
TLDR
It is demonstrated that actively exploring different views of 3D face models produces more robust recognition memory than passively viewing playback of the same moving stimuli and combining multiple depth cues can facilitate recognition performance even though a single depth cue alone rarely produces measurable benefits.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 57 REFERENCES
Face Recognition across Large Viewpoint Changes
TLDR
It is shown that the overlapping visible regions of heads for both surface and texture data can support accurate recognition, even with pose differences of as much as 90 degrees between the learning and test view.
When face recognition fails.
Two studies investigated recognition of pictures of faces, focusing on the effects of changes in appearance of the face from presentation to test and type of processing or encoding. Experiment 1
The basis of the 3/4 view advantage in face recognition
Three-quarter views of faces promote better recognition memory for previously unfamiliar faces than do full-face views. This paper reports experiments which examine the possible basis of the effect,
Face recognition, pose and ecological validity.
This paper describes five experiments with two main aims; first, to assess the effects of variations in photographic facial pose on recognizability. A second, more fundamental aim was to examine the
The importance of symmetry and virtual views in three-dimensional object recognition
TLDR
The results show that the human visual system can indeed exploit symmetry to facilitate object recognition, and support the model for object recognition in which a small number of two-dimensional views are remembered and combined to recognize novel views of the same object.
How are three-dimensional objects represented in the brain?
TLDR
In a variety of psychophysical experiments that explore different aspects of the problem of object recognition and representation in human vision, models based on multiple-views representations replicated many of the psychophysical results concerning the observed pattern of human performance.
Recognition-by-components: a theory of human image understanding.
TLDR
Recognition-by-components (RBC) provides a principled account of the heretofore undecided relation between the classic principles of perceptual organization and pattern recognition.
Face recognition: Effects of study to test maintenance and change of photographic mode and pose
Facial recognition performance was examined as a function of changes in the target between initial exposure (study) and a subsequent recognition test. Photographic mode (colour vs. black and white)
Shape from shading
Understanding how the shape of a three dimensional object may be recovered from shading in a two-dimensional image of the object is one of the most important - and still unresolved - problems in
Recognition by Linear Combinations of Models
  • S. Ullman, R. Basri
  • Mathematics, Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 1991
An approach to visual object recognition in which a 3D object is represented by the linear combination of 2D images of the object is proposed. It is shown that for objects with sharp edges as well as
...
1
2
3
4
5
...