{"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/superquadrics-revisited-learning-3d-shape","title":"Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboids","arxiv_id":"1904.09970","date":"2019-04-22","proceeding":"CVPR 2019 6","authors":["Despoina Paschalidou","Ali Osman Ulusoy","Andreas Geiger"],"abstract":"Abstracting complex 3D shapes with parsimonious part-based representations\nhas been a long standing goal in computer vision. This paper presents a\nlearning-based solution to this problem which goes beyond the traditional 3D\ncuboid representation by exploiting superquadrics as atomic elements. We\ndemonstrate that superquadrics lead to more expressive 3D scene parses while\nbeing easier to learn than 3D cuboid representations. Moreover, we provide an\nanalytical solution to the Chamfer loss which avoids the need for computational\nexpensive reinforcement learning or iterative prediction. Our model learns to\nparse 3D objects into consistent superquadric representations without\nsupervision. Results on various ShapeNet categories as well as the SURREAL\nhuman body dataset demonstrate the flexibility of our model in capturing fine\ndetails and complex poses that could not have been modelled using cuboids.","url_abs":"http://arxiv.org/abs/1904.09970v1","url_pdf":"http://arxiv.org/pdf/1904.09970v1.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":"superquadrics-revisited-learning-3d-shape","repo_url":"https://github.com/paschalidoud/superquadric_parsing","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.09970","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}