Research on how the human brain extracts meaning from sensory input relies in principle on methodological reductionism. In the present study, we adopt a more holistic approach by modeling the cortical responses to semantic information that was extracted from the visual stream of a feature film, employing artificial neural network models. Advances in both computer vision and natural language processing were utilized to extract the semantic representations from the film by combining perceptual and linguistic information. We tested whether these representations were useful in studying the human brain data. To this end, we collected electrocorticography responses to a short movie from 37 subjects and fitted their cortical patterns across multiple regions using the semantic components extracted from film frames. We found that individual semantic components reflected fundamental semantic distinctions in the visual input, such as presence or absence of people, human movement, landscape scenes, human faces, etc. Moreover, each semantic component mapped onto a distinct functional cortical network involving high-level cognitive regions in occipitotemporal, frontal and parietal cortices. The present work demonstrates the potential of the data-driven methods from information processing fields to explain patterns of cortical responses, and contributes to the overall discussion about the encoding of high-level perceptual information in the human brain. Semantic processing of audiovisual material is a topic of great interest in cognitive neuroscience. A lot of knowledge has been accrued with studies focusing on the visual object processing, object categorization and processing of various semantic attributes of the visually perceived objects 1-3. We have come to understand a lot about the ventral stream of visual processing in the brain 4,5 , the object categorization ability of inferior temporal cortex 6-8 as well as the specific roles of fusiform face area 9 , parahippocampal place area 10 , the motion processing temporal region 11 and other areas involved in high-level visual processing 4,12-14. Most studies address the topic with a reductionist approach, with carefully constructed tasks and stimuli to investigate a particular aspect of brain function 15. With new tools available to investigate high-dimensional phenomena, more holistic approaches become feasible. Complex brain mechanisms may be identified or characterized by mapping brain responses onto informational structures that are extracted from non-constrained sensory material. More concretely, one can utilize recent advances in computational modeling in different domains such as computer vision and natural language processing that are driven by extraction of high-level semantic relations directly from the unprocessed input, such as images 16,17 and texts 18-20. Among the most recent and most powerful such advances are deep artificial neural network models. Not only do these models achieve unprecedented performance on solving complex tasks (for example, visual object i...