Title :
Scene Change Adaptation for Scalable Video Coding
Author :
Anselmo, Tea ; Alfonso, Daniele
Author_Institution :
Adv. Syst. Technol. Labs., STMicroelectron., Agrate Brianza, Italy
Abstract :
This document presents a Scene Change Adaptation method for the Scalable Video Coding extension of H.264/AVC. Our method is based on a scene change detection algorithm that identifies transitions in video sequences by using the motion information provided by the pre-analysis phase of a Fast Motion Estimation algorithm based on spatial-temporal motion correlation. Intra coding picture is imposed at the end of the Group Of Pictures containing the scene change, thus dynamically adapting the sequence structure to the scene content and preventing the propagation of the prediction error due to scene change. The proposed algorithm has been also combined with a Buffer-Based Constant Bit-Rate control algorithm, which ensures HRD buffer compliance and target bit-rate requirements. Simulation results show that the SCD algorithm can improve the global coding efficiency and the local visual quality when a scene change occurs.
Keywords :
image sequences; motion estimation; video coding; H.264/AVC; HRD buffer compliance; buffer-based constant bit-rate control algorithm; fast motion estimation algorithm; global coding efficiency; intra coding picture; local visual quality; motion information; prediction error; scalable video coding; scene change adaptation; scene content; sequence structure; spatial-temporal motion correlation; target bit-rate requirements; video sequences; Change detection algorithms; Correlation; Encoding; IP networks; Motion estimation; Streaming media; Video coding;
Conference_Titel :
Signal Processing Conference, 2009 17th European
Conference_Location :
Glasgow
Print_ISBN :
978-161-7388-76-7