The paper describes an object recognition system, based on the co-operation
of several visual modules (early vision, object detector, and object recog
nizer). The system is active because the behavior of each module is tuned o
n the results given by other modules and by the internal models. This solut
ion allows to detect inconsistencies and to generate a feedback process. Th
e proposed strategy has shown good performance especially in case of comple
x scene analysis, and it has been included in the visual system of the DAIS
Y robotics system. Experimental results on real data are also reported.