{"id":15,"date":"2014-11-15T21:38:57","date_gmt":"2014-11-15T19:38:57","guid":{"rendered":"http:\/\/www.mpateraki.org\/?page_id=15"},"modified":"2018-07-27T14:20:13","modified_gmt":"2018-07-27T12:20:13","slug":"research","status":"publish","type":"page","link":"https:\/\/mpateraki.org\/?page_id=15","title":{"rendered":"Research"},"content":{"rendered":"

constantly under construction\u2026<\/p>\n

Datasets<\/h2>\n

Occluded Articulated Human Body Dataset<\/a><\/strong>
\nAn annotated dataset for human body pose extraction and tracking under occlusions.<\/p>\n\n\n\n\n
\"t1\"<\/a><\/td>\n\"t2\"<\/a><\/td>\n<\/tr>\n
\"t3\"<\/a><\/td>\n\"t4\"<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Human Body Pose Estimation<\/h2>\n

A model-based approach for markerless articulated full body pose extraction and tracking in RGB-D sequences. A cylinder-based model is employed to represent the human body. For each body part a set of hypotheses is generated and tracked over time by a Particle Filter. To evaluate each hypothesis, we employ a novel metric that considers the reprojected Top View of the corresponding body part. The latter, in conjunction with depth information, effectively copes with difficult and ambiguous cases, such as severe occlusions.<\/p>\n\n\n\n