The contemporary dispute between different versions of realist philosophies of visual perception is currently at a stage where it is very difficult to give preference to any one of the sides solely on the basis of the arguments presented by the parties. However, one can try to consider this discussion in the wider context of the problem of realism in philosophy in general. One of the most well-founded variants of contemporary scientific realism is ontic structural realism (OSR), within which the concept of a real pattern (RP) has been developed, based on a more detailed analysis of the criteria of reality than simply the idea of the independence of reality from perception. Within OSR, there is no point in arguing about what we see as objects of the real world – things, or their low-level properties, or their high-level properties, since they are all RPs, but seen from different perspectives. Subjectively experienced visual perception can be understood as compressed information about the perceived RP and, accordingly, as a model of this RP. Such models can be different: based on habit, on default justification, on inferences – hence the existence of different realistic philosophies of perception, each of which emphasizes models of a different type.