{"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/modeling-context-between-objects-for","title":"Modeling Context Between Objects for Referring Expression Understanding","arxiv_id":"1608.00525","date":"2016-08-01","proceeding":null,"authors":["Varun K. Nagaraja","Vlad I. Morariu","Larry S. Davis"],"abstract":"Referring expressions usually describe an object using properties of the\nobject and relationships of the object with other objects. We propose a\ntechnique that integrates context between objects to understand referring\nexpressions. Our approach uses an LSTM to learn the probability of a referring\nexpression, with input features from a region and a context region. The context\nregions are discovered using multiple-instance learning (MIL) since annotations\nfor context objects are generally not available for training. We utilize\nmax-margin based MIL objective functions for training the LSTM. Experiments on\nthe Google RefExp and UNC RefExp datasets show that modeling context between\nobjects provides better performance than modeling only object properties. We\nalso qualitatively show that our technique can ground a referring expression to\nits referred region along with the supporting context region.","url_abs":"http://arxiv.org/abs/1608.00525v1","url_pdf":"http://arxiv.org/pdf/1608.00525v1.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":"modeling-context-between-objects-for","repo_url":"https://github.com/varun-nagaraja/referring-expressions","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"object","task_name":"Object"},{"task_slug":"referring-expression","task_name":"Referring Expression"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1608.00525","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}