Bo-Jhih Chen, Shen-Chuan Tai, and Yung-Gi Wu
Zero-quantized block, H.264/Advanced Video Coding, sum of absolute difference, discrete cosine transform, quantization
An efficient Gaussian rate-distortion approach is presented for detecting zero-quantized blocks (ZQBs) in H.264/Advanced Video Coding (H.264/AVC). ZQBs are quite common in low bit-rate video coding. To reduce the number of computations of the integer discrete cosine transform (DCT), quantization (Q), inverse quantization (IQ) and inverse DCT (IDCT), many approaches have been developed to detect ZQBs before DCT. DCT, Q, IQ and IDCT processes are not performed if the input residual data are detected as a ZQB. The proposed Gaussian rate-distortion approach greatly reduces computational cost with negligible video quality degradation. Several test video sequences are used to evaluate performance. Experimental results indicate that the proposed approach reduces the execution time of DCT, Q, IQ and IDCT by an average of 39.45% and obtains an 85.19% detection rate with a reasonable video quality loss of 0.002 dB.
Important Links:
Go Back
