Multimodal 2D and 3D biometrics for face recognition

Results are presented for the largest experimental study to date that investigates the comparison and combination of 2D and 3D face data for biometric recognition. To our knowledge, this is also the only such study to incorporate significant time lapse between gallery and probe image acquisition. Recognition results are presented for gallery and probe datasets of 166 subjects imaged in both 2D and 3D, with six to thirteen weeks time lapse between gallery and probe images of a given subject. Using a PCA-based approach tuned separately for 2D and for 3D, we find no statistically significant difference between the rank-one recognition rates of 83.1% for 2D and 83.7% for 3D. Using a certainty-weighted sum-of-distance approach to combining 2D and 3D, we find a multimodal rank-one recognition rate of 92.8%, which is statistically significantly greater than either 2D or 3D alone.