{"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/what-can-i-do-here-leveraging-deep-3d","title":"What can I do here? Leveraging Deep 3D saliency and geometry for fast and scalable multiple affordance detection","arxiv_id":"1812.00889","date":"2018-12-03","proceeding":null,"authors":["Eduardo Ruiz","Walterio Mayol-Cuevas"],"abstract":"This paper develops and evaluates a novel method that allows for the\ndetection of affordances in a scalable and multiple-instance manner on visually\nrecovered pointclouds. Our approach has many advantages over alternative\nmethods, as it is based on highly parallelizable, one-shot learning that is\nfast in commodity hardware. The approach is hybrid in that it uses a geometric\nrepresentation together with a state-of-the-art deep learning method capable of\nidentifying 3D scene saliency. The geometric component allows for a compact and\nefficient representation, boosting the performance of the deep network\narchitecture which proved insufficient on its own. Moreover, our approach\nallows not only to predict whether an input scene affords or not the\ninteractions, but also the pose of the objects that allow these interactions to\ntake place. Our predictions align well with crowd-sourced human judgment as\nthey are preferred with 87% probability, show high rates of improvement with\nalmost four times (4x) better performance over a deep learning-only baseline\nand are seven times (7x) faster than previous art.","url_abs":"http://arxiv.org/abs/1812.00889v1","url_pdf":"http://arxiv.org/pdf/1812.00889v1.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":"what-can-i-do-here-leveraging-deep-3d","repo_url":"https://github.com/eduard626/interaction-tensor-affordances","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"affordance-detection","task_name":"Affordance Detection"},{"task_slug":"multiple-affordance-detection","task_name":"Multiple Affordance Detection"},{"task_slug":"one-shot-learning","task_name":"One-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}