Rate estimation for CABAC with low complexity
Kwanghyun Won, Byeungwoo Jeon · 2013
The context-adaptive binary arithmetic coding (CABAC) in the H.264/AVC is known to be one of the key technologies to improve coding efficiency. H.264/AVC encoding under rate-distortion optimization demands a lot of computations, especially, in the rate calculation for deciding the best coding mode. In this paper, a fast rate estimation method for CABAC is proposed with its application in mind for such cases of rate-distortion optimized encoding. It estimates the rate utilizing two proposed equations which only refer to the probability state index (pStateIdx) and the indication of whether or not the most probable symbol (MPS) match with the current binary to be coded. Simulation results show that the proposed method achieves run-time saving of about 19.27% (up to 22.46%) on average with no degradation or additional memory requirement.