Hnativ L.O.
Simple high-order integer cosine transform for high-resolution video coding

A matrix method is proposed for constructing a simple order-32 integer cosine transform. Based on the method, a one-norm simple order-32 integer transform is constructed and its fast algorithms of low computational complexity are developed that require only integer operations and whose computational complexity is 4.9 times less than that of well-known algorithms and is 21.6 times less than that of the standard H.265. This transform is close to the discrete cosine transform and has good coding performance. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.

Keywords: AVS, coding gain, compression ratio, computational complexity, discrete cosine transform, factorization, fast transform, H.264, H.265, integer cosine transform, scaled transform, video coding, Codes (symbols), Compression ratio (machinery), Computational complexity, Cosine transforms, Discrete cosine transforms, Factorization, Image coding, Coding gains, Fast Transform, H.264, H.265, Integer cosine transform, Video signal processing

Cite:
Hnativ L.O. (2018). Simple high-order integer cosine transform for high-resolution video coding. Cybernetics and Systems Analysis, 54 (3), 166-177. doi: https://doi.org/10.1007/s10559-018-0050-z http://jnas.nbuv.gov.ua/article/UJRN-0000863968 [In Russian].