An other-race effect for face recognition algorithms

@article{Phillips2011AnOE,
  title={An other-race effect for face recognition algorithms},
  author={P. Jonathon Phillips and Fang Jiang and Abhijit Narvekar and Julianne H. Ayyad and Alice J. O'Toole},
  journal={ACM Trans. Appl. Percept.},
  year={2011},
  volume={8},
  pages={14:1-14:11}
}
Psychological research indicates that humans recognize faces of their own race more accurately than faces of other races. [] Key Result Humans showed the standard other-race effect for these faces, but showed more stable performance than the algorithms over changes in the race of the test faces. State-of-the-art face recognition algorithms, like humans, struggle with “other-race face” recognition.

Figures from this paper

A Control System for Assessing Commercial Face Recognition Software for Racial Bias

A database of still images is created with a total of 132 black-faces representing 22 individuals, having images cropped to pixel sizes of 100x100, 80x80, 60x60 and 40x40 respectively, to provide a control measure for face recognition algorithms.

Accuracy Comparison Across Face Recognition Algorithms: Where Are We on Measuring Race Bias?

It is concluded that race bias needs to be measured for individual applications and a checklist for measuring this bias in face recognition algorithms is provided.

Demographic effects on estimates of automatic face recognition performance

Characterizing the Variability in Face Recognition Accuracy Relative to Race

A methodical investigation into differences in face recognition accuracy between African-American and Caucasian image cohorts of the MORPH dataset finds that, for all four matchers considered, the impostor and the genuine distributions are statistically significantly different between cohorts.

Ingroup and outgroup differences in face detection.

Humans show improved recognition for faces from their own social group relative to faces from another social group. Yet before faces can be recognized, they must first be detected in the visual

Racial Faces in-the-Wild: Reducing Racial Bias by Deep Unsupervised Domain Adaptation

A deep information maximization adaptation network (IMAN) is proposed to bridge the domain gap among different races and the existence of racial bias in FR algorithms, and comprehensive experiments show that the racial bias could be narrowed-down by the algorithm.

Face Age Estimation and the Other-race Effect

The result showed that the age estimation system trained with the predominantly black face database (GA-ANN-AES-855) outperformed the system training with predominantly white faces on testing with the aforementioned black face samples, which established that the improvement in the correct classification rate was statistically significant.

Computational perspectives on the other-race effect

Psychological studies have long shown that human memory is superior for faces of our own-race than for faces of other-races. In this paper, we review computational studies of own- versus other-race
...

References

SHOWING 1-10 OF 47 REFERENCES

An Encoding Advantage for Own-Race versus Other-Race Faces

The results indicate that an own-race advantage occurs at the encoding stage of face processing.

Reversibility of the Other-Race Effect in Face Recognition During Childhood

The testing of adults of Korean origin who were adopted by European Caucasian families when they were between the ages of 3 to 9 indicates that the face recognition system remains plastic enough during childhood to reverse the other-race effect.

Face Recognition Algorithms Surpass Humans Matching Faces Over Changes in Illumination

Seven state-of-the-art face recognition algorithms are compared with humans on a face-matching task and three algorithms surpassed human performance matching face pairs prescreened to be "difficult" and six algorithms surpassed humans on "easy" face pairs.

CLASSIFYING FACES BY RACE : THE STRUCTURE OF FACE CATEGORIES

This article explored the finding that cross-race (CR) faces are more quickly classified by race than same race (SR) faces. T. Valentine and M. Endo (1992) modeled this effect by assuming that face

Structural aspects of face recognition and the other-race effect

The human data and simulation results indicate that the memorahility component of typicality may be related to small, local, distinctive features, whereas the attractiveness/familiarity component may be more related to the global, shape-based properties of the face.

Social and Cognitive Factors Affecting the Own-Race Bias in Whites

This study investigated factors associated with the commonly found own-race bias (ORB) in face recognition. We utilized several measures of general face-recognition memory, visual perception and

The Other-Race Effect Develops During Infancy

The findings suggest that facial input from the infant's visual environment is crucial for shaping the face-processing system early in infancy, resulting in differential recognition accuracy for faces of different races in adulthood.

Race as a visual feature: using visual search and perceptual discrimination tasks to understand face categories and the cross-race recognition deficit.

  • D. Levin
  • Psychology
    Journal of experimental psychology. General
  • 2000
These findings support a new explanation for the cross-race (CR) recognition deficit based on feature coding differences between CR and SR faces, and appear incompatible with similarity-based models of face categories.

Thirty years of investigating the own-race bias in memory for faces: A meta-analytic review

The current article reviews the own-race bias (ORB) phenomenon in memory for human faces, the finding that own-race faces are better remembered when compared with memory for faces of another, less