Training of a classifier

A method and a device for improved training of a classifier. For this purpose, the device for detecting an object has a first image recording device, which is situated in a first position for recording the object from a first perspective. The device furthermore has a stored first classifier, which is configured to detect the object based on the recorded first perspective. At least one second image recording device is situated in a second position for recording the object from a second perspective, which differs from the first perspective. Furthermore, a data processing device is configured to detect the object recorded by the first image recording device based on the first classifier and to assign to the second perspective a class predetermined by the first classifier. This makes it possible to expand the existing classifier to include the detection of the object from different additional perspectives.