{"106721":{"#nid":"106721","#data":{"type":"event","title":"CSE Seminar: Semantic Image Segmentation by Ranking Multiple Figure-Ground Segment Hypotheses","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003ESpeaker:\u0026nbsp;\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cp\u003EFuxin Li (CSE Postdoc),\u0026nbsp;School of Computational Science and Engineering\u003C\/p\u003E\n\u003Cp\u003E\u003Cstrong\u003ETitle:\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cp\u003ESemantic Image Segmentation by Ranking Multiple Figure-Ground Segment Hypotheses\u003C\/p\u003E\n\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cp\u003EThe goal of semantic segmentation is to recognize objects in images and classify each pixel into a particular category or background. It is significantly more challenging than more conventional image classification and object detection problems. Different from most other approaches that mainly rely on local cues from pixels or superpixels, our approach starts from multiple binary segmentations that capture important global cues, such as the shape of an object.\u003C\/p\u003E\n\u003Cp\u003EThese binary segmentations, generated by the unsupervised Constrained Parametric Min-Cut (CPMC) algorithm, cover a spectrum of different locations and segment sizes. In the learning phase, the segmentations are first filtered with a global \u0022objectness\u0022 filter, then fed into a kernel-based learning framework that continuously predicts the overlap of each segmentation with each particular object category. Finally, cues from multiple highly-ranked segmentations are used to determine the classification of each pixel. From 2009 to 2011, this approach has won the prestigious PASCAL VOC Segmentation Challenge three times in a row.\u003C\/p\u003E\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"CSE Seminar: Semantic Image Segmentation by Ranking Multiple Figure-Ground Segment Hypotheses"}],"uid":"27592","created_gmt":"2012-02-06 14:34:11","changed_gmt":"2016-10-08 01:57:52","author":"Joshua Preston","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2012-02-10T13:00:00-05:00","event_time_end":"2012-02-10T14:00:00-05:00","event_time_end_last":"2012-02-10T14:00:00-05:00","gmt_time_start":"2012-02-10 18:00:00","gmt_time_end":"2012-02-10 19:00:00","gmt_time_end_last":"2012-02-10 19:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"1304","name":"High Performance Computing (HPC)"},{"id":"47223","name":"College of Computing"},{"id":"50877","name":"School of Computational Science and Engineering"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022mailto:lebanon@cc.gatech.edu\u0022\u003EDr. Guy Lebanon\u003C\/a\u003E\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}