3-D wavelet video coding technique provides an elegant solution to the multimedia transmission scalability requirement. It uses motion-compensated temporal filtering (MCTF) for temporal decomposition and 2-D discrete wavelet transform (2-D DWT) for spatial decomposition. MCTF decomposes video frames into temporal high-pass residuals and temporal low-pass residuals. The temporal high-pass residuals have similar statistical properties as the motion compensation residuals but the temporal low-pass ones are similar to the original images. The adaptive directional lifting-based wavelet transform (ADLWT) finds the best filter direction in an image local area and provides better energy compaction than DWT. In this paper, we use ADLWT to replace 2-D DWT to improve the performance of 3-D wavelet video coding.
|Number of pages||4|
|State||Published - 1 Dec 2011|
|Event||Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011, APSIPA ASC 2011 - Xi'an, China|
Duration: 18 Oct 2011 → 21 Oct 2011
|Conference||Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011, APSIPA ASC 2011|
|Period||18/10/11 → 21/10/11|