Image Segmentation for Improved Lossless Screen Content Compression

Uddehal SR, Strutz T, Och H, Kaup A (2023)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2023

Publisher: IEEE

City/Town: ieee.org

Pages Range: 5

Event location: Rhode Islands, Greece GR

URI: https://ieeexplore.ieee.org/document/10094983

DOI: 10.1109/ICASSP49357.2023.10094983

Abstract

In recent years, it has been found that screen content images (SCI) can be effectively compressed based on appropriate probability modelling and suitable entropy coding methods such as arithmetic coding. The key objective is determining the best probability distribution for each pixel position. This strategy works particularly well for images with synthetic (textual) content. However, usually screen content images not only consist of synthetic but also pictorial (natural) regions. These images require diverse models of probability distributions to be optimally compressed. One way to achieve this goal is to separate synthetic and natural regions. This paper proposes a segmentation method that identifies natural regions enabling better adaptive treatment. It supplements a compression method known as Soft Context Formation (SCF) and operates as a pre-processing step. If at least one natural segment is found within the SCI, it is split into two subimages (natural and synthetic parts) and the process of modelling and coding is performed separately for both. For SCIs with natural regions, the proposed method achieves a bit-rate reduction of up to 11.6% and 1.52% with respect to HEVC and the previous version of the SCF

Authors with CRIS profile

Shabhrish Reddy Uddehal Lehrstuhl für Multimediakommunikation und Signalverarbeitung Hannah Och Lehrstuhl für Multimediakommunikation und Signalverarbeitung Andre Kaup Lehrstuhl für Multimediakommunikation und Signalverarbeitung

Involved external institutions

Hochschule für angewandte Wissenschaften Coburg DE Germany (DE)

How to cite

APA:

Uddehal, S.R., Strutz, T., Och, H., & Kaup, A. (2023). Image Segmentation for Improved Lossless Screen Content Compression. In Proceedings of the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 5). Rhode Islands, Greece, GR: ieee.org: IEEE.

MLA:

Uddehal, Shabhrish Reddy, et al. "Image Segmentation for Improved Lossless Screen Content Compression." Proceedings of the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, Rhode Islands, Greece ieee.org: IEEE, 2023. 5.

BibTeX: Download