Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
In this paper, an efficient H.264/AVC baseline encoder, described in RVC-CAL actor language, is introduced. The main aim of the paper is two folds: a) to demonstrate the flexibility and ease that is provided by RVC-CAL, which allows for efficient implementation of the presented encoder, and b) to shed light on the advantages that can be brought into the RVC framework by including such encoding tools. The main modules of the designed encoder include: Inter Frame Prediction (Motion Estimation/Compensation), Intra Frame Prediction, and Entropy Coding. Descriptions of the designed modules, accompanied with RVC-CAL design issues are provided. A comparison between different development approaches is also provided. The obtained results show that specifying complex video codecs (e.g. H.264/AVC encoder) using RVC-CAL followed by automatic translation into HDL, which is achievable by the tools that support the standard, results in more efficient HW implementation compared to the traditional HW design flow. A discussion that explains the reasons behind such results concludes the paper.
Sabine Süsstrunk, Yufan Ren, Peter Arpad Grönquist, Alessio Verardo, Qingyi He
, ,
David Atienza Alonso, Marina Zapater Sancho, Arman Iranfar, William Andrew Simon