Combined Spatial Correlation of Coded Neighboring Blocks for Intra-frame Coding Mode Prediction
Yu Han, Aidong Men, Ziyi Quan
Abstract
Yu Han, Aidong Men, Ziyi Quan
Abstract
A novel algorithm for coding the intra 4times4 prediction mode selection information with combined spatial correlation of coded neighboring blocks is proposed in this work. A large number of bits are used to transmit the prediction modes of each intra coded block in the H.264/AVC standard, and these bits cost becomes more sensitive in the case of low bit rate application. The proposed algorithm utilizing an L-shaped template to predict and code the mode selection information has significant improvement of compression efficiency. It is proposed to be applied to the next generation video compression algorithm. For some test sequence, the proposed algorithm can achieve 12.62% bit rate reduction compared to the state-of-the-art H.264/AVC standard at the expense of 0.219 dB PSNR degradation.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A novel algorithm for coding the intra 4times4 prediction mode selection information with combined spatial correlation of coded neighboring blocks is proposed in this work. A large number of bits are used to transmit the prediction modes of each intra coded block in the H.264/AVC standard, and these bits cost becomes more sensitive in the case of low bit rate application. The proposed algorithm utilizing an L-shaped template to predict and code the mode selection information has significant improvement of compression efficiency. It is proposed to be applied to the next generation video compression algorithm. For some test sequence, the proposed algorithm can achieve 12.62% bit rate reduction compared to the state-of-the-art H.264/AVC standard at the expense of 0.219 dB PSNR degradation.
Key concepts: Computer science, Algorithm, Coding (social sciences), Bit rate, Context-adaptive binary arithmetic coding, Data compression, Spatial correlation, Real-time computing