Recognising Faces: Effects of Lighting Direction, Inversion, and Brightness Reversal

Abstract

When information about three-dimensional shape obtained from shading and shadows is ambiguous, the visual system favours an interpretation of surface geometry which is consistent with illumination from above. If pictures of top-lit faces are rotated the resulting stimulus is both figurally inverted and illuminated from below. In this study the question of whether the effects of figural inversion and lighting orientation on face recognition are independent or interactive is addressed. Although there was a clear inversion effect for faces illuminated from the front and above, the inversion effect was found to be reduced or eliminated for faces illuminated from below. A strong inversion effect for photographic negatives was also found but in this case the effect was not dependent on the direction of illumination. These findings are interpreted as evidence to suggest that lighting faces from below disrupts the formation of surface-based representations of facial shape.

Eye contrast polarity is critical for face recognition by infants

“…This finding of the poorer face recognition performance for negatives is consistent with findings for adult participants (e.g. Bruce & Langton, 1994;Johnston, et al, 1992). …”

“…Anstis, 2005;Bruce & Langton, 1994;Johnston, Hill, Carman, 1992;Kemp, McManus, & Pigott, 1990;Lewis & Johnston, 1997). This effect has been attributed to the unusual pigmentation (Bruce & Langton, 1994;Russel, Sinha, Biederman, & Nederhouser, 2006) and/or the unnatural pattern of shading interfering with three-dimensional face perception (Johnston, Hill, Carman 1992;Liu, Collin, Burton, & Chaudhuri, 1999).…”

Recovering Facial Shape Using a Statistical Model of Surface Normal Direction

“…This potentially has some interesting psychological implications. Typically, poor human performance, when perceiving images of faces illuminated from below, has been attributed to an illumination direction estimation process in which the light source is constrained to come from above [20]. However, with perfect knowledge of the illumination direction, our algorithm recovers less accurate facial shape information when the illumination is from below compared to above.…”

To what extent can matching algorithms based on direct outputs of spatial filters account for human object recognition?

“…The high correlation between human psychophysical data in judging facial identity and the performance of the FRS found in Kalocsai, et al (1994) suggests that the human visual system might exploit these advantages for face recognition. However, this strategy comes at a price: human face recognition performance is drastically disrupted by changes in the filter values produced by changes in orientation, direction of contrast, lighting, and rotation in depth (Bruce, Valentine, & Baddeley, 1987;Johnston, Hill, & Carman, 1992;Kalocsai et al, 1994).…”