Computational & Technology Resources
an online resource for computational,
engineering & technology publications
engineering & technology publications
Civil-Comp Conferences
ISSN 2753-3239
ISSN 2753-3239
CCC: 12
PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GPU AND CLOUD COMPUTING FOR ENGINEERING
PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GPU AND CLOUD COMPUTING FOR ENGINEERING
Edited by: P. Iványi, J. Kruis and B.H.V. Topping
Paper 2.7
Performance Constraints in IME-HEVC Software Integration
O.M. López-Granado1, M. Martínez-Rach1, H. Migallón1, R. Gutierrez Mazon2 and M. Perez Malumbres1
1Computer Engineering Department, Miguel Hernández University, Elche, Spain
2Communications Engineering Department, Miguel Hernández University, Elche, Spain
Full Bibliographic Reference for this paper
O.M. López-Granado, M. Martínez-Rach, H. Migallón, R. Gutierrez Mazon, M. Perez Malumbres, "Performance Constraints in IME-HEVC Software Integration", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Eighth International Conference on
Parallel, Distributed, GPU and Cloud Computing for Engineering", Civil-Comp Press, Edinburgh, UK,
Online volume: CCC 12, Paper 2.7, 2025,
Keywords: video codecs, HEVC, FPGA, IME, temporal prediction, Sum of Absolute Difference architecture.
Abstract
High-Efficiency Video Coding (HEVC) improves compression efficiency, compared to H.264/AVC, but motion estimation remains a major computational bottleneck, especially for high-resolution video sequences. To accelerate encoding, we implemented a hardware-based motion estimation module, achieving significant speed-ups. However, integrating this hardware with software introduces constraints that may limit its impact on overall encoding performance.
This paper analyzes these integration challenges, identifying key bottlenecks in the software/hardware encoding process. Based on this analysis, we propose a refined integration approach combining the hardware module with a slice-based parallel HEVC encoder. This optimized version achieves a significant speed-up of up to 146.90 times compared to the sequential HEVC encoder using full-search motion estimation.
download the full-text of this paper (PDF, 9 pages, 253 Kb)
go to the previous paper
return to the table of contents
return to the volume description