University of Thessaly (UTH)
Parallel and content-adaptive video coding and transcoding for next-generation standards
Abstract
dc:descriptionThe continuous demand for high-quality video services, driven by the widespread adoption of ultra-high-definition content, real-time streaming, and advanced multimedia applications, has made efficient video compression and transcoding a problem of significant scientific and technological importance. Modern video coding standards, including High Efficiency Video Coding (HEVC), AOMedia Video 1 (AV1), and Versatile Video Coding (VVC), achieve remarkable compression performance through the introduction of increasingly sophisticated coding tools. However, these improvements are accompanied by a substantial increase in computational complexity, limiting real-time processing and the efficient utilization of modern computing resources. Motivated by these challenges, this thesis investigates methods for improving the computational efficiency of modern video coding and transcoding systems without compromising coding performance. The proposed research investigates parallel video coding, prediction-driven optimization, content-adaptive coding, and heterogeneous transcoding as the main directions for improving computational efficiency. More specifically, the impact of tile partitioning on the coding efficiency of AV1 is experimentally evaluated, heterogeneous scheduling strategies for wavefront and tile-based parallelism are investigated, and a deep learning framework is proposed for predicting computational bottlenecks during parallel video coding. Furthermore, image segmentation techniques are exploited to derive content-aware subpicture partitions for the VVC encoder, while a fast HEVC-to-VVC transcoding methodology is developed by reusing coding information extracted from the input bitstream. Experimental results demonstrate that efficient parallelization, machine learning, content-aware optimization, and coding information reuse can substantially reduce computational complexity while maintaining high coding efficiency. Overall, the proposed methodologies contribute to the development of scalable and computationally efficient video coding and transcoding systems, facilitating the practical deployment of next-generation video compression technologies in modern multimedia applications.
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Panagou, Natalia
- Πανάγου, Ναταλία
Subjects
dc:subject × 25- Κωδικοποίηση βίντεο
- Διακωδικοποίηση βίντεο
- Παράλληλη επεξεργασία
- Προσαρμοστική κωδικοποίηση βίντεο
- Συμπίεση βίντεο
- Αλγόριθμοι και υπολογιστική πολυπλοκότητα
- Βελτιστοποίηση αλγορίθμων
- Αποδοτικότητα κωδικοποίησης
- Video coding
- Video transcoding
- Parallel processing
- Content-adaptive video coding
- Video compression
- Computational complexity
- Algorithm optimization
- Coding efficiency
- High efficiency video coding (HEVC)
- Versatile video coding (VVC)
- AOMedia Video 1 (AV1)
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Επιστήμη ηλεκτρονικών υπολογιστών
- Computer and Information Sciences
- Natural Sciences
- Computer Science
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/62972