首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Region based foreground segmentation combining color and depth sensors via logarithmic opinion pool decision
Institution:1. College of Information Science and Technology, Beijing Normal University, China;2. Lawrence Berkeley National Laboratory, United States;3. College of Information Science and Technology, Agriculture University of Hebei, China;4. School of Computer and Information Technology, Beijing Jiaotong University, China;5. School of Civil Engineering, Beijing Jiaotong University, China;1. Palaeontology, Geobiology and Earth Archives Research Centre, School of Biological, Earth and Environmental Sciences, UNSW, Kensington, NSW 2052, Australia;2. School of Humanities and Languages, UNSW, Kensington, NSW 2052, Australia;3. Royal Botanic Gardens Victoria, Melbourne, VIC 3004, Australia;4. Pharmacoinformatics Laboratory, Discipline of Pharmacology, Bosch Institute and Sydney Medical School, The University of Sydney, NSW, Australia;1. Department of Hepatobiliary Surgery, First Affiliated Hospital of Medical College, Xi’an Jiaotong University, 277 West Yanta Road, Xi’an 710061, Shaanxi Province, China;2. Department of General Surgery, First Affiliated Hospital of Medical College, Xi’an Jiaotong University, 277 West Yanta Road, Xi’an 710061, Shaanxi Province, China;3. Department of Oncology, First Affiliated Hospital of Medical College, Xi’an Jiaotong University, 277 West Yanta Road, Xi’an 710061, Shaanxi Province, China;4. Department of Pharmaceutical Sciences, North Dakota State University, Fargo, USA;1. Department of Hepatobiliary Surgery, First Affiliated Hospital Medical College of Xi’an Jiaotong University, Xi’an, Shaanxi 710061, China;2. Department of General Surgery, First Affiliated Hospital Medical College of Xi’an Jiaotong University, Xi’an, Shaanxi 710061, China
Abstract:In this paper we present a novel foreground segmentation system that combines color and depth sensors information to perform a more complete Bayesian segmentation between foreground and background classes. The system shows a combination of spatial-color and spatial-depth region-based models for the foreground as well as color and depth pixel-wise models for the background in a Logarithmic Opinion Pool decision framework used to correctly combine the likelihoods of each model. A posterior enhancement step based on a trimap analysis is also proposed in order to correct the precision errors that the depth sensor introduces. The results presented in this paper show that our system is robust in front of color and depth camouflage problems between the foreground object and the background, and also improves the segmentation in the area of the objects’ contours by reducing the false positive detections that appear due to the lack of precision of the depth sensors.
Keywords:Foreground segmentation  Space-color models  Space-depth models  Kinect camera  GMM  Color and depth combination  Logarithmic Opinion Pool  Hellinger distance
本文献已被 ScienceDirect 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号