{"656072":{"#nid":"656072","#data":{"type":"event","title":"PhD Defense by John Lambert","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003ETitle:\u003C\/strong\u003E Deep Learning for Building and Validating Geometric and Semantic Maps\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cbr \/\u003E\r\n\u003Cstrong\u003EDate:\u003C\/strong\u003E Monday, March 21, 2022\u003Cbr \/\u003E\r\n\u003Cstrong\u003ETime:\u003C\/strong\u003E 4 pm (EDT)\u003Cbr \/\u003E\r\n\u003Cstrong\u003ELocation (Virtual):\u003C\/strong\u003E\u0026nbsp;\u003Ca href=\u0022https:\/\/gatech.zoom.us\/j\/99561135585?pwd=VDFCREYyeDNMTEN5RC9OREVCREhlQT09\u0022\u003Ehttps:\/\/gatech.zoom.us\/j\/99561135585?pwd=VDFCREYyeDNMTEN5RC9OREVCREhlQT09\u003C\/a\u003E\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\nJohn Lambert\u003Cbr \/\u003E\r\nPh.D. Student in Computer Science\u003Cbr \/\u003E\r\nSchool of Interactive Computing, College of Computing\u003Cbr \/\u003E\r\nGeorgia Institute of Technology\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\n\u003Cstrong\u003ECommittee:\u003C\/strong\u003E\u003Cbr \/\u003E\r\nDr. James Hays (Co-Advisor), School of Interactive Computing, Georgia Institute of Technology\u003Cbr \/\u003E\r\nDr. Frank Dellaert (Co-Advisor), School of Interactive Computing, Georgia Institute of Technology\u003Cbr \/\u003E\r\nDr. Zsolt Kira, School of Interactive Computing, Georgia Institute of Technology\u003Cbr \/\u003E\r\nDr. Cedric Pradalier, School of Interactive Computing, Georgia Institute of Technology, Lorraine\u003Cbr \/\u003E\r\nDr. Simon Lucey, Australian Institute for Machine Learning, University of Adelaide\u003Cbr \/\u003E\r\n\u003Cbr \/\u003E\r\n\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EMapping the world is an essential tool for making spatial artificial intelligence a reality in our near future. Spatial AI, or embodied intelligence for 3D perception, enables awareness and understanding of our surroundings. Current methods for building and validating geometric and semantic maps are limited in several ways. For example, floorplan maps constructed from sparse camera views within indoor environments generally suffer from low completeness. In other domains, such as city streets, the world is ever-changing, making online validation of high-definition (HD) maps a requirement for today\u0026#39;s self-driving vehicles; however, many current methods suffer from high-storage costs or limited accuracy.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn this dissertation, I first develop a new learning-based approach, SALVe, for creating complete and accurate 2d geometric maps (floorplans) under very wide baselines and occlusion. Second, I explore the role of the\u0026nbsp;deep \u0026quot;front end\u0026quot; in Structure-from-Motion (SfM), and analyze its use in\u0026nbsp;a new system for global SfM, GTSFM. Finally, I introduce learning-based formulations for solving the HD map change detection task in a bird\u0026rsquo;s eye view and ego-view. We collect the first public dataset for the task, which we entitle the Trust, but Verify (TbV) dataset, by mining thousands of hours of data from over 9 months of autonomous vehicle fleet operations. Because real map changes are infrequent and vector maps are easy to synthetically manipulate, we lean on simulated\u0026nbsp;data to train such models. Perhaps surprisingly, we show that such models can generalize to real world distributions.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Deep Learning for Building and Validating Geometric and Semantic Maps"}],"uid":"27707","created_gmt":"2022-03-07 12:53:27","changed_gmt":"2022-03-07 12:53:27","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2022-03-21T17:00:00-04:00","event_time_end":"2022-03-21T19:00:00-04:00","event_time_end_last":"2022-03-21T19:00:00-04:00","gmt_time_start":"2022-03-21 21:00:00","gmt_time_end":"2022-03-21 23:00:00","gmt_time_end_last":"2022-03-21 23:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}