Title :
Unified loop filter for high-performance video coding
Author :
Liu, Yu ; Huo, Yan
Author_Institution :
Hong Kong Appl. Sci. & Technol. Res. Inst. (ASTRI), Shatin, China
Abstract :
In this paper, we propose a unified loop filter for high-performance video coding, which suppresses the quantization noise optimally and improves the objective and subjective quality of the reconstructed picture simultaneously. The proposed filter unifies nonlinear enhancement filter (for removing blocking and ringing artifacts) and linear restoration filter (for improving coding efficiency) within the classical optimization framework of least mean square error (LMSE). Experimental results show that the proposed unified loop filter achieves on average 11.53% objective coding gain in combination with better visual quality improvement around edges and textures, compared with H.264/AVC High Profile.
Keywords :
filtering theory; image enhancement; image restoration; least mean squares methods; optimisation; quantisation (signal); video coding; classical optimization framework; coding efficiency improvement; high performance video coding; least mean square error; linear restoration filter; nonlinear enhancement filter; quantization noise; unified loop filter; visual quality improvement; Automatic voltage control; Encoding; Mean square error methods; Optimization; Pixel; Quantization; Vectors; Unified loop filter; least mean square error; linear restoration filter; nonlinear enhancement filter;
Conference_Titel :
Multimedia and Expo (ICME), 2010 IEEE International Conference on
Conference_Location :
Suntec City
Print_ISBN :
978-1-4244-7491-2
DOI :
10.1109/ICME.2010.5582600