{"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/tags2parts-discovering-semantic-regions-from","title":"Tags2Parts: Discovering Semantic Regions from Shape Tags","arxiv_id":"1708.06673","date":"2017-08-22","proceeding":"CVPR 2018 6","authors":["Sanjeev Muralikrishnan","Vladimir G. Kim","Siddhartha Chaudhuri"],"abstract":"We propose a novel method for discovering shape regions that strongly\ncorrelate with user-prescribed tags. For example, given a collection of chairs\ntagged as either \"has armrest\" or \"lacks armrest\", our system correctly\nhighlights the armrest regions as the main distinctive parts between the two\nchair types. To obtain point-wise predictions from shape-wise tags we develop a\nnovel neural network architecture that is trained with tag classification loss,\nbut is designed to rely on segmentation to predict the tag. Our network is\ninspired by U-Net, but we replicate shallow U structures several times with new\nskip connections and pooling layers, and call the resulting architecture\n\"WU-Net\". We test our method on segmentation benchmarks and show that even with\nweak supervision of whole shape tags, our method can infer meaningful semantic\nregions, without ever observing shape segmentations. Further, once trained, the\nmodel can process shapes for which the tag is entirely unknown. As a bonus, our\narchitecture is directly operational under full supervision and performs\nstrongly on standard benchmarks. We validate our method through experiments\nwith many variant architectures and prior baselines, and demonstrate several\napplications.","url_abs":"http://arxiv.org/abs/1708.06673v3","url_pdf":"http://arxiv.org/pdf/1708.06673v3.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":"tags2parts-discovering-semantic-regions-from","repo_url":"https://github.com/sanjeevmk/Tags2Parts","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"tag","task_name":"TAG"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"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}