Fast matching pursuit with vector norm comparison

Matching pursuit was demonstrated useful especially in low-bit-rate video coding. However, its massive computation required for finding atoms hinders its use from practical application. This paper is to provide a new method that can drastically reduce the computational load without any degradation in performance of matching pursuit. We compare vector norms based on the Schwarz inequality to preclude substantial number of dictionary functions without actually evaluating their inner products in atom search. Experimental results show that the number of inner product calculation is only about 20%-35% of the conventional separability-based fast method.