Face Recognition from Correspondence Maps
Rolf P. Würtz · 2022
This chapter presents several algorithms that have proved useful in the task of recognizing human faces from gray-value images without additional information. Their common feature is the extraction of correspondence maps between an image to be analyzed and several stored views of known objects, which are called models. The processing of a retinal gray-level image in simple cells of the primary visual cortex can be modeled by a wavelet transform based on complex-valued Gabor functions. The major drawback of correspondence-based recognition systems is that the computationally expensive procedure of creating a correspondence map must be carried out for each of the stored models. Humans do very poorly in recognizing faces from photographic negatives. One recognition system has no problem with this, because Gabor amplitudes are identical for positive and negative images. There is now increasing evidence that the human brain employs different modules for the recognition of faces and general objects.