Low dimensional search for efficient texture synthesis

Low dimensional search for efficient texture synthesis

Show other versions (1)

Current texture synthesis algorithms rely upon high dimensional approximate nearest neighbour (ANN) searches to determine the best pixel to use at the current position. The feature vectors used in the ANN search are typically between 100 and 300 dimensions. A large amount of research has examined ho...

Full description

Bibliographic Details
Main Author: Kimberley, Fred
Language:English
Published: 2009
Online Access:http://hdl.handle.net/2429/15561
id ndltd-LACETR-oai-collectionscanada.gc.ca-BVAU.2429-15561
record_format oai_dc
spelling ndltd-LACETR-oai-collectionscanada.gc.ca-BVAU.2429-155612014-03-14T15:48:18Z Low dimensional search for efficient texture synthesis Kimberley, Fred Current texture synthesis algorithms rely upon high dimensional approximate nearest neighbour (ANN) searches to determine the best pixel to use at the current position. The feature vectors used in the ANN search are typically between 100 and 300 dimensions. A large amount of research has examined how to reduce the number of feature vectors that need to be searched but very little has been done to speed up the actual comparisons. We present two new texture synthesis algorithms that use an order of magnitude fewer dimensions during the ANN search. In addition, we construct and make use of an error texture that further reduces the time spent comparing two feature vectors. 2009-11-23T21:55:55Z 2009-11-23T21:55:55Z 2004 2009-11-23T21:55:55Z 2004-11 Electronic Thesis or Dissertation http://hdl.handle.net/2429/15561 eng UBC Retrospective Theses Digitization Project [http://www.library.ubc.ca/archives/retro_theses/]
collection NDLTD
language English
sources NDLTD
description Current texture synthesis algorithms rely upon high dimensional approximate nearest neighbour (ANN) searches to determine the best pixel to use at the current position. The feature vectors used in the ANN search are typically between 100 and 300 dimensions. A large amount of research has examined how to reduce the number of feature vectors that need to be searched but very little has been done to speed up the actual comparisons. We present two new texture synthesis algorithms that use an order of magnitude fewer dimensions during the ANN search. In addition, we construct and make use of an error texture that further reduces the time spent comparing two feature vectors.
author Kimberley, Fred
spellingShingle Kimberley, Fred
Low dimensional search for efficient texture synthesis
author_facet Kimberley, Fred
author_sort Kimberley, Fred
title Low dimensional search for efficient texture synthesis
title_short Low dimensional search for efficient texture synthesis
title_full Low dimensional search for efficient texture synthesis
title_fullStr Low dimensional search for efficient texture synthesis
title_full_unstemmed Low dimensional search for efficient texture synthesis
title_sort low dimensional search for efficient texture synthesis
publishDate 2009
url http://hdl.handle.net/2429/15561
work_keys_str_mv AT kimberleyfred lowdimensionalsearchforefficienttexturesynthesis
_version_ 1716653327799812096

Similar Items