Abstract: Primates are very good at recognizing objects and at the same time assessing viewing angle and relative position to the object. In such tasks they outperform existing computer vision systems. We present a model for the unsupervised learning of codes for object identities and for position and viewing angle (including depth rotation), where each code is independent of all others. The model is based on a hierarchy of Slow Feature Analysis (SFA) modules, which were shown to be a good model for complex cells in the early visual system. A similar architecture has previously been shown to model hippocampal place cells, head direction cells and spatial-view cells from naturalistic videos. Here, our model is trained and tested with visual input data generated from virtual 3D-objects. For cases with few objects, the model produces independent codes for object identity and pose in a completely unsupervised way. To show that the model extracts useful information also in more complex scenarios we use linear regressions (for position and viewing angle) or simple classifiers (for object identity). This supervised step extracts the feature codes with high invariance under the applied object transformations of translation, rotation and zoom.