Face recognition using the POEM descriptor

Real-world face recognition systems require careful balancing of three concerns: computational cost, robustness, and discriminative power. In this paper we describe a new descriptor, POEM (patterns of oriented edge magnitudes), by applying a self-similarity based structure on oriented magnitudes and prove that it addresses all three criteria. Experimental results on the FERET database show that POEM outperforms other descriptors when used with nearest neighbour classifiers. With the LFW database by combining POEM with GMMs and with multi-kernel SVMs, we achieve comparable results to the state of the art. Impressively, POEM is around 20 times faster than Gabor-based methods.

► POEM is built by computing self-similarity of magnitudes over different directions.
► POEM is robust and distinctive (Section 3.2).
► Recognition results on FERET show that POEM outperforms other descriptors (Table 2).
► On LFW, by using POEM and simple classifiers we get the state of the art (Table 3).
► Our descriptor leads to a low complexity system (Section 3.4).