Publication Details
Comparison of light field compression methods
Šolony Marek, Ing., PhD. (DCGM FIT BUT)
Chlubna Tomáš, Ing. (DCGM FIT BUT)
Dlabaja Drahomír, Ing. (FIT BUT)
Klíma Ondřej, Ing., Ph.D. (DCGM FIT BUT)
Zemčík Pavel, prof. Dr. Ing. (DCGM FIT BUT)
4D light fields, Plenoptic imaging, Compression, Image refocusing, 3D Reconstruction
In this article, we compare the impact of state-of-the-art light field compression methods. It addresses quality of (a) refocused images and (b) point clouds reconstructed from 4D light field data. The methods include recent video compression formats, specifically H.265, AV1, XVC, and H.266/VVC (finalized in 2020). In addition, we have extended a standard image compression method into four dimensions and compared it with the video compression formats. It turned out that the new VVC format demonstrated superior performance, closely followed by the underrated XVC. Apart from the comparison, we show that the four-dimensional light field data can be compressed with a higher ratio than independent still images while maintaining the same visual quality of a perceived picture.
@ARTICLE{FITPUB12613,
author = "David Ba\v{r}ina and Marek \v{S}olony and Tom\'{a}\v{s} Chlubna and Drahom\'{i}r Dlabaja and Ond\v{r}ej Kl\'{i}ma and Pavel Zem\v{c}\'{i}k",
title = "Comparison of light field compression methods",
pages = "2517--2528",
journal = "Multimedia Tools and Applications",
volume = 81,
number = 2,
year = 2022,
ISSN = "1573-7721",
doi = "10.1007/s11042-021-11645-x",
language = "english",
url = "https://www.fit.vut.cz/research/publication/12613"
}