3D-HEVC Fast CCL Intra Partitioning Algorithm For Low Bitrate Applications

Three-Dimensional High-Efficiency Video Coding (3D-HEVC) has transformed the transmission and storage of stereoscopic and multi-view video content by efficiently encoding texture and depth information. Depth map coding is essential for synthesised views which create the 3D effect. However faces chal...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Multi-Conference on Systems, Signals, and Devices s. 749 - 755
Hlavní autoři: Omran, Nacir, Kachouri, Rostom, Maraoui, Amna, Werda, Imen, Belgacem, Hamdi
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 17.02.2025
Témata:
3D-HEVC
Bit rate
Computational efficiency
Connected Component Labeling
CTU Partitioning
Depth Maps
Encoding
Intra Coding
Labeling
Quality assessment
Stereo image processing
Streaming media
Three-dimensional displays
Video compression
ISSN:2474-0446
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Three-Dimensional High-Efficiency Video Coding (3D-HEVC) has transformed the transmission and storage of stereoscopic and multi-view video content by efficiently encoding texture and depth information. Depth map coding is essential for synthesised views which create the 3D effect. However faces challenges in meeting the growing demand for low-bitrate encoding with faster processing times in applications like cloud streaming, low-latency broadcasting and 3D video editing software. The Coding Tree Unit (CTU) partitioning, crucial for coding efficiency, is particularly impactful for depth maps due to their sharp edges and large homogeneous regions. To address this problem, we propose a Connected Component Labeling (CCL) approach to optimise CTU partitioning for depth map coding in intra-prediction for QP pairs QP 35-42 and 40-45. By leveraging the structural properties of depth maps, CCL enables precise partitioning, thus reducing unnecessary computations and improving compression efficiency. Our method targets applications requiring efficient and fast low-bitrate encoding. Experimental results show a great reduction in encoding time with 70.64% while slightly increasing bitrate by 0.08% and managing minimal quality degradation.
ISSN:2474-0446
DOI:10.1109/SSD64182.2025.10989985