An evaluation of multimodal 2D+3D face biometrics

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 27; no. 4; pp. 619 - 624
• Main Authors: Chang, K.I., Bowyer, K.W., Flynn, P.J.
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01-04-2005; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Biometrics; Biometry - methods; Cluster Analysis; Computer science; control theory; systems; Digital cameras; Exact sciences and technology; Face - anatomy & histology; Face recognition; Female; Galleries; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image recognition; Image sensors; Imaging, Three-Dimensional - methods; Index Terms- Biometrics; Information Storage and Retrieval - methods; Lenses; Male; multimodal; multisample; Pattern analysis; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Probes; Recognition; Reproducibility of Results; Sensitivity and Specificity; Speech; Subtraction Technique; Terminology; Three dimensional; three-dimensional face; Two dimensional; Weighting; Biometrics; Image recognition; Face recognition; three-dimensional face; Tridimensional image; Multimodal interface; multimodal; Multiple sampling; multisample
• ISSN: 0162-8828; 1939-3539; 2160-9292
• DOI: 10.1109/TPAMI.2005.70

We report on the largest experimental study to date in multimodal 2D+3D face recognition, involving 198 persons in the gallery and either 198 or 670 time-lapse probe images. PCA-based methods are used separately for each modality and match scores in the separate face spaces are combined for multimodal recognition. Major conclusions are: 1) 2D and 3D have similar recognition performance when considered individually, 2) combining 2D and 3D results using a simple weighting scheme outperforms either 2D or 3D alone, 3) combining results from two or more 2D images using a similar weighting scheme also outperforms a single 2D image, and 4) combined 2D+3D outperforms the multi-image 2D result. This is the first (so far, only) work to present such an experimental control to substantiate multimodal performance improvement.