A multi-step alignment scheme for face recognition in range images

Abstract

Face Recognition in range images is a challenging task, especially if the pose of the shown face is unknown. To solve this, an alignment procedure consisting of facial feature hypotheses extraction by invariant curvature features, PCA-based classification and Iterative Closest Point alignment will be introduced to create aligned and normalized patches. These patches will then be used in a recognition algorithm, a discrete Pseudo 2-Dimensional Hidden Markov Model approach based on vector quantized DCTmod2 features. The results of this processing chain are discussed and compared to previous works.

An Efficiently 3D Face Recognizing Method Using Range Image and Sparse Representation

“…After the registration, we can obtain better corresponding region in range image. Unlike statistical method which requires accurate pre-processing because it is sensitive to noise [10], sparse representation classification method dose not belong to statistical method, so it is not necessary to carry out the refine alignment in the range images domain considering the time-consuming hausdorff matching process described in [10] even this step can promise better recognition result.…”