可調式運算量在H.264/AVC的壓縮標準中提出了可變區塊大小的編碼模式。這種新技術可以有效率地壓縮影片，卻也增加了編碼器的計算複雜度。但是過於複雜的編碼器，不容易實做在實際用應用中。在過去已經有許多加速模式選擇技術被提出以解決計算量過於複雜的問題，但這些方法都只適用特定平台和需要特定計算量。由於在不同平台上的應用，計算能力也會有所不同。因此，我們希望能設計一套加速編碼模式演算法，該演算法能根據不同計算能力得到對應的壓縮品質。
我們提出了利用優先權序列(priority list)上的特徵來決定區塊的編碼模式的三階漸進式演算法(3 Stage Progressive Algorithm )。這裡的優先權序列(priority list)是利用（0,0）位置的Mean Square Error (MSE)排序後形成。根據統計的結果，優先權序列與區塊編碼模式有以下的關係：
1. 較複雜的編碼模式容易聚集在優先權序列的前半部，而較簡單的則在序列後半部。
2. 比較相鄰兩張圖片(frame)畫面上相同位置的區塊在前後兩個優先權序列上優先順序的變化。1）在相鄰序列上位置變化小於10％時，這些區塊的編碼模式會跟前一張圖片的區塊類似；2）在相鄰兩在序列上位置變化大於10％時，如果優先順序變化提升，這些區塊大多是較複雜的編碼模式，反之，則為較簡單的編碼模式。
鑑於以上觀察，及為了達到可調節式運動量的需求，我們將編碼模式選擇演算法分為三個階段。第一階段：根據區塊在優先權序列位置指定編碼模式，在序列前端的區塊使用8×8後端使用16×16，並只用四個4×4子區塊做向量估計。第二階段：根據區塊在優先權序列的順序變化指定編碼模式，順序變化小的採用前一張圖片的編碼模式，變化大的則依變化方向決定，並只用八個4×4子區塊做向量估計。第三階段：根據第一階段到第二階段改進的結果，對有機會進一步提升品質的區塊採取進一步的複雜編碼模式，進行16×16的區塊向量估計。若前兩階段無顯著改進，則採進簡單編碼模式進行16×16的區塊向量估計。
實驗結果顯示我們的方法可以有效加速原本H.264/AVC向量估計和模式選擇的速度。在計算量上，我們可在1/2到1/64倍之間調整向量估計演算法的複雜度，提供向量估計模組2到19倍的加速倍率調整能力，而整體的編碼時間則有1.3到3.5倍的加速倍率調整空間。儘管隨著加速倍率的提高，壓縮品質的損失仍能控制在可接受的範圍內。

H.264/AVC introduces variable-block size coding modes, which can improve coding efficiency but also increases computational complexity very much. The extreme computational complexity makes H.264/AVC hard to be realized practically. In the past, many fast mode decision algorithms were proposed to partially reduce the complexity of mode selection. However, most of them are with fixed complexity. In this study we attempt to bring out a fast mode decision algorithm for H.264/AVC adapting to different computation limits.
A mode decision algorithm based on computation-aware architecture is proposed. This algorithm uses the characteristics of a priority list to make a decision on coding modes based on some profitable observations on the relationship between priority lists, corresponding to consecutive frames and their block coding modes.
The simulation results show that the proposed method can scale down the H.264/AVC complexity from 1/2 to 1/64 in terms of the number of 4×4 block SADs required, while maintaining the acceptable video quality. In real implementation on H.264/AVC JM9, this program has 2 to 19 times speed-up in motion estimation time, or 1.3 to 3.5 times speed-up of the total encoding time.

[1] ITU-T, “Video codec for audiovisual services at px64 kbits/s,” ITU-T Rec. H.261 v1: Nov 1990, v2: Mar. 1993.
[2] ISO/IEC JTC 1, “Coding of moving pictures and associated audio for digital storage media at up to about 1.5 Mbit/s – Part 2: Video,” ISO/IEC 11172 (MPEG-1), Nov. 1993.
[3] ITU-T and ISO/IEC JTC 1, “Generic coding of moving pictures and associated audio information – Part 2: Video,” ITU-T Rec. H.262 and ISO/IEC 13818-2 (MPEG-2), Nov. 1994 (with several subsequent amendments and corrigenda).
[4] ITU-T, “Video coding for low bit rate communication,” ITU-T Rec. H.263; v1: Nov. 1995, v2: Jan. 1998, v3: Nov. 2000.
[5] ISO/IEC JTC 1, “Coding of audio-visual objects – Part 2: Visual,” ISO/IEC 14496-2 (MPEG-4 Part 2), Jan. 1999 (with several subsequent amendments and corrigenda).
[6] ITU-T, “Video coding for low bit rate communication,” ITU-T Rec. H.263; v1: Nov. 1995, v2: Jan. 1998, v3: Nov. 2000.
[7] “ITU-T Rec.H.264, ISO/IEC 14496-10 AVC, “Advanced Video Coding Final Draft International Standard”, JVT-G050r1, Geneva, Switzerland,” May 2003.
[8] JVT reference software, JM 9.3, download from http://bs.hhi.de/~suehring/tml
[9] T. Koga, K. Iinuma, A. Hirano, Y. Iijima, and T. Ishiguro, “Motion compensated interframe coding for video conferencing,” Proc. Nat. Telecommun. Conf., New Orleans, LA, pp. G5.3.1-G5.3.5, Dec 1981.
[10] R. Li, B. Zeng, and M.L. Liou, “A new three-step search algorithm for block motion estimation,” IEEE Trans. on Circuits and Systems for Video Technology, vol. 4, no. 4, pp. 438-42, Aug 1994.
[11] S. Zhu and K.K. Ma, “A new diamond search algorithm for fast block matching motion estimation,” Proc. of Int. Conf. Information, Communications and Signal Processing, vol.1, pp.292-6, 1997.
[12] J.Y. Tham, S. Ranganath, M. Ranganath, and A.A. Kassim, “A novel unrestricted center-biased diamond search algorithm for block motion estimation,” IEEE Trans. On Circuits & Systems for Video Technology, vol.8, pp.369-77, Aug. 1998.
[13] P.L Tai, S.Y. Huang, C.T. Liu, and T.S. Wang, “Computation-aware scheme for Software-based block motion estimation,” IEEE Trans. On Circuits & Systems for Video Technology, vol.13, pp.901-913, Sep. 2003.
[14] P.L. Tai, C.T. Liu, and J.S. Wang, “Complexity-adaptive search algorithm for block motion estimation,” in ICIP 2001, vol.2, pp.969 - 972 Oct. 2001.
[15] X. Jing and L.P. Chau, “An efficient inter mode decision approach for H.264 video coding,” in ICME 2004, vol.2, pp. 1111 - 1114 June 2004.
[16] D. Zhu, Q. Dai and R. Ding, “Fast inter prediction mode decision for H.264,” in ICME 2004, vol.2, pp. 1123 – 1126, June 2004.
[17] A.C. Yu, “Efficient block-size selection algorithm for inter-frame coding in H.264/MPEG-4 AVC,” in ICASSP 2004, vol.3, pp. 169-72, May 2004.
[18] M. Yang and W. Wang, “Fast macroblock mode selection based on motion content classification in H.264/AVC,” in ICIP 2004, vol.2, pp. 741 - 744, Oct. 2004.
[19] C.D. Bei and R.M. Gray, “An improvement of the minimum distortion encoding algorithm for vector quantization,” IEEE Trans. Communication, vol. COM-33, pp.1132-1133, Oct. 1985.