For a consistent analysis of a visual scene the different features of an individual object have to be recognized as belonging together and separated from other objects. In the brain of higher vertebrates it has been suggested that this is achieved by synchronization of the activity of disjunct nerve cell assemblies. However cross-talk between spatially adjacent image parts occurs preventing efficient synchronization. As a consequence image segmentation can be heavily impaired in network models. In these cases, temporal differences, naturally introduced by stimulus latencies in every biological sensory system, can strongly improve the performance of the network.
展开▼