Developments in digital cinema and computer games industry have required for fast texture synthesis algorithms that are capable of synthesizing better quality textures. Unfortunately most of the existing state-of-the-art algorithms suffer from excessive complexity related issues resulting in high synthesis times. This thesis proposes seven novel algorithms in order to address the existing problems in sample-based texture synthesis. All the proposed algorithms use visually significant transform coefficients within texture synthesis and thus effectively exploit the limitations of the human visual system (HVS). This thesis proposes four algorithms that use the Discrete Wavelet Transform (DWT) for extracting visually significant information within the texture synthesis process.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:421921 |
| Date | January 2005 |
| Creators | Wickramanayake, Dhammike Saranath |
| Publisher | Loughborough University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | https://dspace.lboro.ac.uk/2134/34552 |
Page generated in 0.0019 seconds