Using the structure and motion of stereo point clouds for the semantic segmentation of images

The segmentation of images into semantically coherent regions has been approached in many different ways in the over 40 years since the problem was first addressed. Recently systems using the motion of point clouds derived from laser depth scanners and structure from motion have been described, but...

Full description

Bibliographic Details
Main Author: Dockrey, Matthew
Language:English
Published: University of British Columbia 2010
Online Access:http://hdl.handle.net/2429/17415
id ndltd-UBC-oai-circle.library.ubc.ca-2429-17415
record_format oai_dc
spelling ndltd-UBC-oai-circle.library.ubc.ca-2429-174152018-01-05T17:24:00Z Using the structure and motion of stereo point clouds for the semantic segmentation of images Dockrey, Matthew The segmentation of images into semantically coherent regions has been approached in many different ways in the over 40 years since the problem was first addressed. Recently systems using the motion of point clouds derived from laser depth scanners and structure from motion have been described, but these are monetarily and computationally expensive options. We explore the use of stereo cameras to achieve the same results. This approach is shown to work in an indoor environment, giving results that compare favorably with existing systems. The use of stereo instead of structure from motion is shown to be preferable in this environment, while the choice of stereo algorithm proves highly critical to the quality of the results. The use of aggregated voting regions is explored, which is shown to moderately improve the results while speeding up the process considerably. Experiments are also run biasing the randomized input to the classifier generation process, which show further improvements in both performance and execution time. Overall, the approach is shown to be feasible, but not currently practical for robotic navigation in this environment. Science, Faculty of Computer Science, Department of Graduate 2010-01-04T16:37:35Z 2010-01-04T16:37:35Z 2009 2010-05 Text Thesis/Dissertation http://hdl.handle.net/2429/17415 eng Attribution-ShareAlike 3.0 Unported http://creativecommons.org/licenses/by-sa/3.0/ University of British Columbia
collection NDLTD
language English
sources NDLTD
description The segmentation of images into semantically coherent regions has been approached in many different ways in the over 40 years since the problem was first addressed. Recently systems using the motion of point clouds derived from laser depth scanners and structure from motion have been described, but these are monetarily and computationally expensive options. We explore the use of stereo cameras to achieve the same results. This approach is shown to work in an indoor environment, giving results that compare favorably with existing systems. The use of stereo instead of structure from motion is shown to be preferable in this environment, while the choice of stereo algorithm proves highly critical to the quality of the results. The use of aggregated voting regions is explored, which is shown to moderately improve the results while speeding up the process considerably. Experiments are also run biasing the randomized input to the classifier generation process, which show further improvements in both performance and execution time. Overall, the approach is shown to be feasible, but not currently practical for robotic navigation in this environment. === Science, Faculty of === Computer Science, Department of === Graduate
author Dockrey, Matthew
spellingShingle Dockrey, Matthew
Using the structure and motion of stereo point clouds for the semantic segmentation of images
author_facet Dockrey, Matthew
author_sort Dockrey, Matthew
title Using the structure and motion of stereo point clouds for the semantic segmentation of images
title_short Using the structure and motion of stereo point clouds for the semantic segmentation of images
title_full Using the structure and motion of stereo point clouds for the semantic segmentation of images
title_fullStr Using the structure and motion of stereo point clouds for the semantic segmentation of images
title_full_unstemmed Using the structure and motion of stereo point clouds for the semantic segmentation of images
title_sort using the structure and motion of stereo point clouds for the semantic segmentation of images
publisher University of British Columbia
publishDate 2010
url http://hdl.handle.net/2429/17415
work_keys_str_mv AT dockreymatthew usingthestructureandmotionofstereopointcloudsforthesemanticsegmentationofimages
_version_ 1718582300044689408