{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-a-discriminative-model-for-the","title":"Learning a Discriminative Model for the Perception of Realism in Composite Images","arxiv_id":"1510.00477","date":"2015-10-02","proceeding":"ICCV 2015 12","authors":["Jun-Yan Zhu","Philipp Krähenbühl","Eli Shechtman","Alexei A. Efros"],"abstract":"What makes an image appear realistic? In this work, we are answering this\nquestion from a data-driven perspective by learning the perception of visual\nrealism directly from large amounts of data. In particular, we train a\nConvolutional Neural Network (CNN) model that distinguishes natural photographs\nfrom automatically generated composite images. The model learns to predict\nvisual realism of a scene in terms of color, lighting and texture\ncompatibility, without any human annotations pertaining to it. Our model\noutperforms previous works that rely on hand-crafted heuristics, for the task\nof classifying realistic vs. unrealistic photos. Furthermore, we apply our\nlearned model to compute optimal parameters of a compositing method, to\nmaximize the visual realism score predicted by our CNN model. We demonstrate\nits advantage against existing methods via a human perception study.","url_abs":"http://arxiv.org/abs/1510.00477v1","url_pdf":"http://arxiv.org/pdf/1510.00477v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-a-discriminative-model-for-the","repo_url":"https://github.com/junyanz/RealismCNN","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1510.00477","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}