H.264 是近年來備受矚目的新視訊壓縮標準，其在移動估測的地方加入了許多新的技術，比如說可變區塊模式大小、四分之一像素精確度，以及多幅參考畫面技術等等，這些新的技術造就了H.264 有更強大的壓縮率以及效能，不過也因此增加了龐大的運算複雜度。有鑑於此，本篇論文提出了『可選擇性多幅參考畫面移動估測演算法』，簡稱SMRFME，我們利用理論推導證明來探討畫面的穩定特性，並以參考畫面的消費成本比較來判斷目前區塊是否需要繼續參考下一張畫面，此外參考第一張畫面結束時便已經有機制判定是否需要繼續參考其他畫面。 另外我們還提出了兩個提前結束參考畫面移動估測的方法，並將其加入SMRFME 中，使運算複雜度可以降得更低。最後提出的『可選擇性多幅參考畫面移動估測演算法』和AFMFSA 演算法[15]做一個比較，結果證明在各方面本論文的演算法都勝過AFMFSA 許多。由此可知本論文提出的演算法在節省時間表現傑出，且保持一定水準的畫面品質與效能。 H.264/AVC features many advanced techniques in motion estimation, such as multiple reference frames, variable block sizes, and quarter pixel resolution. These new features result in significant improvement in coding performance, but with extremely high computational complexity. To improve computation efficiency, we suggest a selective multiple reference frames motion estimation (SMRFME). We use the stationary property to check whether a mode or a block is necessary to perform motion estimation on next reference frame. Additionally, in the proposed SMRFME the cost function referred to the nearest frame is used to skip unnecessary motion estimation on other reference frames. Two early stop criteria are employed to further lessen the computation cost. Finally, our performance comparison is also made with the AFMFSA algorithm [15], and the experiment results demonstrate that the SMRFME outperforms the distinct algorithm in all aspects.
