{"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/affordancenet-an-end-to-end-deep-learning","title":"AffordanceNet: An End-to-End Deep Learning Approach for Object Affordance Detection","arxiv_id":"1709.07326","date":"2017-09-21","proceeding":null,"authors":["Thanh-Toan Do","Anh Nguyen","Ian Reid"],"abstract":"We propose AffordanceNet, a new deep learning approach to simultaneously\ndetect multiple objects and their affordances from RGB images. Our\nAffordanceNet has two branches: an object detection branch to localize and\nclassify the object, and an affordance detection branch to assign each pixel in\nthe object to its most probable affordance label. The proposed framework\nemploys three key components for effectively handling the multiclass problem in\nthe affordance mask: a sequence of deconvolutional layers, a robust resizing\nstrategy, and a multi-task loss function. The experimental results on the\npublic datasets show that our AffordanceNet outperforms recent state-of-the-art\nmethods by a fair margin, while its end-to-end architecture allows the\ninference at the speed of 150ms per image. This makes our AffordanceNet well\nsuitable for real-time robotic applications. Furthermore, we demonstrate the\neffectiveness of AffordanceNet in different testing environments and in real\nrobotic applications. The source code is available at\nhttps://github.com/nqanh/affordance-net","url_abs":"http://arxiv.org/abs/1709.07326v3","url_pdf":"http://arxiv.org/pdf/1709.07326v3.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":"affordancenet-an-end-to-end-deep-learning","repo_url":"https://github.com/nqanh/affordance-net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"affordancenet-an-end-to-end-deep-learning","repo_url":"https://github.com/beapc18/AffordanceNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"affordance-detection","task_name":"Affordance Detection"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.07326","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}