Title : 
Multi-lambda search for improved rate-distortion optimization of H.265/HEVC
         
        
            Author : 
Sio-Kei Im;Ka-Hou Chan
         
        
            Author_Institution : 
MPI-QMUL Information Systems Research Centre, Macao Polytechnic Institute, Macao, China
         
        
        
        
        
            Abstract : 
Lagrangian Rate Distortion Optimization (RDO) is widely employed in modern video encoders such as H.265/HEVC for mode decision. For each encoding stage, this approach involves minimizing a cost, which is a function of rate, distortion and a multiplier called lambda. This paper explores multi-lambda search and its impact on RDO by using more than one value of lambda. Simulations show an average bit rate saving up to 13.0% can be achieved for the H.265/HEVC codec. Moreover, the extra added complexity is contained to a certain level and is also adjustable according the processing resources available.
         
        
            Keywords : 
"Encoding","Rate-distortion","Motion estimation","Optimization","Complexity theory","Bit rate","Distortion"
         
        
        
            Conference_Titel : 
Information, Communications and Signal Processing (ICICS), 2015 10th International Conference on
         
        
        
            DOI : 
10.1109/ICICS.2015.7459952