From image sequences to natural language: a First step toward automatic perception and description of motions

The problem of automatic description of a natural scene is usually divided into two steps: (1) constructing an abstract propositional description of the scene a task carried out by a vision system, and (2) expressing these abstract descriptions in natural language – a task carried out by a naturallanguagesystem. Both of these tasks have been dealt with separately. Systems like MORIO try to solve the first step while others, like HAM-ANS or NAOS deal with the second, using simulated data of the scene description.