29 January 2007 An area-efficient VLSI architecture for AVS intra frame encoder
Author Affiliations +
In this paper, we propose a VLSI architecture for AVS intra frame encoder. Reconstruction loop hinders the parallelism exploration and becomes the critical path in an intra frame encoder. A First Selection Then Prediction (FSTP) method is proposed to break the loop and enable the parallel process of intra mode selection and reconstruction on neighboring blocks. In addition, area-efficient modules were developed. Configurable intra predictor can support all the intra prediction modes. A CA-2D-VLC engine with an area-efficient Exp-Golomb encoder was developed to meet the encoding speed demand with comparably low hardware cost. Synthesized with 0.18 m CMOS standard-cell library, the overall hardware cost of the proposed intra frame encoder is 89k logic gates at the clock frequency constraint of 125MHz. Proposed encoder can satisfy real time encoding of 720x576 4:2:0 25fps video at the working frequency of 54MHz.
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ke Zhang, Lu Yu, "An area-efficient VLSI architecture for AVS intra frame encoder", Proc. SPIE 6508, Visual Communications and Image Processing 2007, 650822 (29 January 2007); doi: 10.1117/12.703475; https://doi.org/10.1117/12.703475

Back to Top