Aprendizado de Máquina Aplicado à Redução do Custo Computacional do Test Zone Search e da Estimação de Movimento Affine do Codificador VVC

• Ramiro Gomes da Silva Viana UFPel
• Luciano Volcan Agostini UFPel

DOI: https://doi.org/10.5753/webmedia_estendido.2025.16358

Resumo

As the demand for video transmission surges in remote work, education, and streaming services, the need for continuous advancements in video encoding technologies becomes increasingly evident. Adapting to the evolving requirements of efficient video delivery and consumption, Versatile Video Coding (VVC) emerges as the state-of-the-art. This dissertation focuses on two key encoding tools within VVC Inter-Frame prediction: Test Zone Search (TZS) and Affine Motion Estimation (AME). A fast TZS and AME approach using Machine Learning is presented, specifically employing Decision Trees. The proposed approach achieved an average reduction of over 20% in total VVC encoding time while maintaining less than a 1% impact on BD-BR coding efficiency.

Palavras-chave: Versatile Video Coding, Test Zone Search, Estimação de Movimento Affine, Árvores de Decisão, Redução de Custo Computacional

Referências

Frank Bossen, Jill Boyce, Xiang Li, Vadim Seregin, and Karsten Suhring. 2020. VTM Common Test Conditions and Software Reference Configurations for SDR Video. [link]

Benjamin Bross, Jianle Chen, Jens-Rainer Ohm, Gary J. Sullivan, and Ye-KuiWang. 2021. Developments in International Video Coding Standardization After AVC, With an Overview of Versatile Video Coding (VVC). Proc. IEEE 109, 9 (2021), 1463–1493. DOI: 10.1109/JPROC.2020.3043399

CISCO. 2023. Cisco Annual Internet Report (2018–2023). [link]

Rafael Ferreira, Luciano Agostini, Cláudio M. Diniz, and Bruno Zatt. 2023. Evaluation of Imprecise Subtractors into Test Zone Search for VVC Encoding. In 2023 36th SBC/SBMicro/IEEE/ACM Symposium on Integrated Circuits and Systems Design (SBCCI). 1–6. DOI: 10.1109/SBCCI60457.2023.10261956

Karsten Suehring. 2022. VTM-16.2. [link]

Ramiro Gomes da Silva Viana. 2025. Acelerando o Algoritmo de Estimação de Movimento Affine do Padrão VVC Utilizando Árvores de Decisão. Dissertação de Mestrado. Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brasil. [link] Programa de Pós-Graduação em Computação (PPGC).