{"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/semantic-instance-labeling-leveraging","title":"Semantic Instance Labeling Leveraging Hierarchical Segmentation","arxiv_id":"1708.00946","date":"2017-08-02","proceeding":null,"authors":["Steven Hickson","Irfan Essa","Henrik Christensen"],"abstract":"Most of the approaches for indoor RGBD semantic la- beling focus on using\npixels or superpixels to train a classi- fier. In this paper, we implement a\nhigher level segmentation using a hierarchy of superpixels to obtain a better\nsegmen- tation for training our classifier. By focusing on meaningful segments\nthat conform more directly to objects, regardless of size, we train a random\nforest of decision trees as a clas- sifier using simple features such as the 3D\nsize, LAB color histogram, width, height, and shape as specified by a his-\ntogram of surface normals. We test our method on the NYU V2 depth dataset, a\nchallenging dataset of cluttered indoor environments. Our experiments using the\nNYU V2 depth dataset show that our method achieves state of the art re- sults\non both a general semantic labeling introduced by the dataset (floor,\nstructure, furniture, and objects) and a more object specific semantic\nlabeling. We show that training a classifier on a segmentation from a hierarchy\nof super pixels yields better results than training directly on super pixels,\npatches, or pixels as in previous work.","url_abs":"http://arxiv.org/abs/1708.00946v1","url_pdf":"http://arxiv.org/pdf/1708.00946v1.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":"semantic-instance-labeling-leveraging","repo_url":"https://github.com/StevenHickson/3DSceneClassification","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"superpixels","task_name":"Superpixels"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}