The paper discusses the problem of visual recognition of several hand postures and gestures. 
Processing is performed in a top-view scenario with a top-mounted camera that monitors the 
user's hands on the working desktop. For postures recognition the chosen classifier was 
a multilayered Perceptron with three layers. By considering an appropriate control of the environment, 
we obtain a real time 25 fps functional system with high detection and recognition accuracy results.  
The interpretation of gestures may be done by estimating also the kinematics parameters. To the end, 
is presented a Video Information System, that manages a database with a multi-level index structure 
for video spatio-temporal knowledge. An SQL operators are defined for searching in this database.