{"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/compositional-gan-learning-conditional-image","title":"Compositional GAN: Learning Image-Conditional Binary Composition","arxiv_id":"1807.07560","date":"2018-07-19","proceeding":null,"authors":["Samaneh Azadi","Deepak Pathak","Sayna Ebrahimi","Trevor Darrell"],"abstract":"Generative Adversarial Networks (GANs) can produce images of remarkable\ncomplexity and realism but are generally structured to sample from a single\nlatent source ignoring the explicit spatial interaction between multiple\nentities that could be present in a scene. Capturing such complex interactions\nbetween different objects in the world, including their relative scaling,\nspatial layout, occlusion, or viewpoint transformation is a challenging\nproblem. In this work, we propose a novel self-consistent\nComposition-by-Decomposition (CoDe) network to compose a pair of objects. Given\nobject images from two distinct distributions, our model can generate a\nrealistic composite image from their joint distribution following the texture\nand shape of the input objects. We evaluate our approach through qualitative\nexperiments and user evaluations. Our results indicate that the learned model\ncaptures potential interactions between the two object domains, and generates\nrealistic composed scenes at test time.","url_abs":"http://arxiv.org/abs/1807.07560v3","url_pdf":"http://arxiv.org/pdf/1807.07560v3.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":"compositional-gan-learning-conditional-image","repo_url":"https://github.com/azadis/CompositionalGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.07560","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}