Abstract
dc:descriptionVideo encoding techniques are getting increasingly important due to the fast-growing video application demands worldwide, which require high-quality videos coupled with low bit rates for storage and streaming efficiency. Silky-smooth video streaming with high frame rates at high resolutions is an evolving research area for the video encoding ecosystem. As the necessity of supporting ever-increasing demands in video resolution leads to new video coding standards, the challenge of harnessing their computational overhead becomes important. Such overhead stems not only from the increased image data due to higher resolutions but also from the coding techniques per se, that are introduced by each standard to improve compression. All modern standards in the field of video coding offer high compression efficiency but this is achieved as aforementioned, by increasing the computational complexity of both the encoding and decoding parts. Ultra-High-Definition (UHD) videos, bring new encoding implementation patterns, that are being recommended for parallelization schemes. Many popular video standards, are sharing different parallelization approaches. In this thesis research, we developed algorithms to tackle the aforementioned encoding complexity through CPU parallelism and/or GP-GPU (General Purpose-Graphics Processing Units) exploitation. We focus on a multithreaded approach for coarse-grained parallel video encoding, that is based on an entirely new encoding scheme and evaluate the performance gains as well as the visual quality of the algorithms under various scenarios to effectively minimize the computational load. As an extension of the previous efforts, we engaged the video graphics card, to investigate the mapping of a novel GPU-specific fast search motion estimation algorithm, the resulting challenges and benefits therein. Finally, we added a fraction execution resolver to avoid unnecessary accuracy calculations for the motion estimation process and to further improve the computational efficiency.
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Papaioannou, George
- Παπαϊωάννου, Γεώργιος
Subjects
dc:subject × 9- Κωδικοποίηση video
- Video coding
- Optimization algorithms
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Υπολογιστές, Υλικό (hardware) και Αρχιτεκτονική
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Computer science, Hardware and Architecture
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/56240