{"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/scenenet-understanding-real-world-indoor","title":"SceneNet: Understanding Real World Indoor Scenes With Synthetic Data","arxiv_id":"1511.07041","date":"2015-11-22","proceeding":null,"authors":["Ankur Handa","Viorica Patraucean","Vijay Badrinarayanan","Simon Stent","Roberto Cipolla"],"abstract":"Scene understanding is a prerequisite to many high level tasks for any\nautomated intelligent machine operating in real world environments. Recent\nattempts with supervised learning have shown promise in this direction but also\nhighlighted the need for enormous quantity of supervised data --- performance\nincreases in proportion to the amount of data used. However, this quickly\nbecomes prohibitive when considering the manual labour needed to collect such\ndata. In this work, we focus our attention on depth based semantic per-pixel\nlabelling as a scene understanding problem and show the potential of computer\ngraphics to generate virtually unlimited labelled data from synthetic 3D\nscenes. By carefully synthesizing training data with appropriate noise models\nwe show comparable performance to state-of-the-art RGBD systems on NYUv2\ndataset despite using only depth data as input and set a benchmark on\ndepth-based segmentation on SUN RGB-D dataset. Additionally, we offer a route\nto generating synthesized frame or video data, and understanding of different\nfactors influencing performance gains.","url_abs":"http://arxiv.org/abs/1511.07041v2","url_pdf":"http://arxiv.org/pdf/1511.07041v2.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":"scenenet-understanding-real-world-indoor","repo_url":"https://github.com/ankurhanda/SceneNetv1.0","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"scene-understanding","task_name":"Scene Understanding"}],"methods":[],"datasets_introduced":[{"slug":"scenenet","name":"SceneNet","full_name":"SceneNet"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1511.07041","atlas_url":"https://app.syntology.ai/?focus=1511.07041","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}