A Model of Saliency-Based Visual Attention for Rapid Scene Analysis
IEEE Transactions on Pattern Analysis and Machine Intelligence
Slow and Smooth: A Bayesian theory for the combination of local motion signals in human vision
Slow and Smooth: A Bayesian theory for the combination of local motion signals in human vision
| Hi-index | 0.00 |
The Selective Attention Scheme has attracted renowned interest in the field of sensorimotor control and visual recognition problems. Especially, selective attention is crucial in terms of saving computational cost for constructing a sensorimotor control system, as the amount of sensory inputs over the system far exceeds its information processing capacity. In fact, selective attention plays an integral role in sensory information processing, enhancing neuronal responses to important or task-relevant stimuli at the expense of the neuronal responses to irrelevant stimuli. To compute human selective attention scheme, we assume that each attention modeled as a probabilistic class must correctly be learned to yield the relationship with different sensory inputs by learning schemes in the first place (sensory modality). Afterwards, their learned probabilistic attention classes can straightforwardly be used for the control property of selecting attention (shifting attention). In this paper, the soundness of proposed human selective attention scheme has been shown in particular with perceptual sensory modality. The scheme is actually realized by a neural network, namely PFC-based network.