A comparison of local versus global imagedecompositions for visual speechreadingMichael

What is the appropriate spatial scale for image representation? In the primate visual system, receptive elds are small at early stages of processing (area V1), and larger at late stages of processing (areas MT, IT). In the current work, we explore the eeciency of local and global image representations on an automatic visual speech recognition task using an HMM as the recognition system. We compare local and global principal component and independent component image representations for the task. Local representations consistently and signiicantly outperformed global representations in terms of generalization to new speakers.