Lossy Light Field Compression Using Modern Deep Learning and Domain Randomization Techniques

Simple item page
dc.contributor.advisorWu, Jianhong
dc.contributor.authorValtchev, Svetozar Jeliazkov
dc.date.accessioned2022-12-14T16:24:15Z
dc.date.available2022-12-14T16:24:15Z
dc.date.copyright2022-06-14
dc.date.issued2022-12-14
dc.date.updated2022-12-14T16:24:13Z
dc.degree.disciplineMathematics & Statistics
dc.degree.levelDoctoral
dc.degree.namePhD - Doctor of Philosophy
dc.description.abstractLossy data compression is a particular type of informational encoding utilizing approximations in order to efficiently tradeoff accuracy in favour of smaller file sizes. The transmission and storage of images is a typical example of this in the modern digital world. However the reconstructed images often suffer from degradation and display observable visual artifacts. Convolutional Neural Networks have garnered much attention in all corners of Computer Vision, including the tasks of image compression and artifact reduction. We study how lossy compression can be extended to higher dimensional images with varying viewpoints, known as light fields. Domain Randomization is explored in detail, and used to generate the largest light field dataset we are aware of, to be used as training data. We formulate the task of compression under the frameworks of neural networks and calculate a quantization tensor for the 4-D Discrete Cosine Transform coefficients of the light fields. In order to accurately train the network, a high degree approximation to the rounding operation is introduced. In addition, we present a multi-resolution convolutional-based light field enhancer, producing average gains of 0.854 db in Peak Signal-to-Noise Ratio, and 0.0338 in Structural Similarity Index Measure over the base model, across a wide range of bitrates.
dc.identifier.urihttp://hdl.handle.net/10315/40642
dc.languageen
dc.rightsAuthor owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.
dc.subjectApplied mathematics
dc.subjectArtificial intelligence
dc.subject.keywordsCompression
dc.subject.keywordsLight field
dc.subject.keywordsDomain randomization
dc.subject.keywordsDomain transfer
dc.subject.keywordsSynthetic data
dc.subject.keywordsPlenoptic
dc.subject.keywordsLossy
dc.subject.keywordsNeural network
dc.subject.keywordsDCT
dc.subject.keywordsImage enhancement
dc.subject.keywordsCompression artifacts
dc.subject.keywordsArtifact reduction
dc.subject.keywordsJPEG
dc.titleLossy Light Field Compression Using Modern Deep Learning and Domain Randomization Techniques
dc.typeElectronic Thesis or Dissertation

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Valtchev_Svetozar_J_2022_Phd.pdf
Size:
54.41 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
license.txt
Size:
1.87 KB
Format:
Plain Text
Description:
Download
No Thumbnail Available
Name:
YorkU_ETDlicense.txt
Size:
3.39 KB
Format:
Plain Text
Description:
Download

Collections

Mathematics & Statistics

All items in the YorkSpace institutional repository are protected by copyright, with all rights reserved except where explicitly noted.