Title :
Full floor identification in images with minimal close range 3D information
Author :
Smith, Graeme ; Morley, J.
Author_Institution :
Horizon Digital Econ. Res., Univ. of Nottingham, Nottingham, UK
fDate :
Sept. 30 2012-Oct. 3 2012
Abstract :
In this paper we propose a method for identifying full floor free space within single images from 3D consumer devices that does not make any assumptions with regard to room geometry and is relatively robust to texture and clutter. Importantly, we show that the method performs well even with very limited depth information showing potential for use with even the lower range consumer devices for which depth information is only available to around 2 metres.
Keywords :
floors; image segmentation; image texture; 3D consumer devices; clutter robustness; depth information; full floor free space identification method; image segmentation; minimal close range 3D information; texture robustness; Cameras; Floors; Geometry; Image color analysis; Image edge detection; Image segmentation; Robustness; 3D processing; Floor identification; Segmentation;
Conference_Titel :
Image Processing (ICIP), 2012 19th IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
978-1-4673-2534-9
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2012.6467038