隨著影像處理技術的蓬勃發展，影片插補成為了其中之一的關鍵技術，為了執行影片插補的技術而可大略定義出兩個步驟去完成。第一步是關鍵影格的選取而第二步則是影格的插補。在本篇論文中提出一個新的關鍵影格選取機制，結合快速二維8×8離散餘弦轉換和逆轉換演算法以加速應用於影片插補的三維離散餘弦轉換演算法。此用在關鍵影格抽取的新方法是基於視覺上的內容將影片中的每一張影格切分成子區塊以計算兩張相鄰影格位於相同位置上的子區塊之相似性差異。一組關鍵影格是整合位在影片裡每張影格相同位置上子區塊的差異性所選取具代表性的影格而成。此組關鍵影格是準備用於影格插補的步驟中，運用三維離散餘弦轉換演算法以插補關鍵影格集合裡相鄰兩張關鍵影格中的影格。於實驗結果中會將採用快速二維離散餘弦轉換和逆轉換演算法加快插補的執行速度，與區塊關鍵影格選取機制以偵測影片中小部分所造成的變化這些優點做一個呈現。

With the rapid growth of development in image processing technologies, the video frame interpolation techniques become one of the key parts. To implement the video interpolation technique, it can be roughly defined two steps to accomplish it. The first step is key frame extraction and the second is frame interpolation. In this thesis, a new key frame selection mechanism together with the fast two-dimensional (2-D) 8×8 forward and inverse discrete cosine transform (DCT) algorithms, which are used to accelerate the three-dimensional (3-D) DCT algorithm, is used to implement the video interpolation technique. The new method utilized to extract the key frames is based on the visual content and each frame of the video sequence is divided into sub-blocks for computing the similarity differences between corresponding sub-blocks of two adjacent frames. A key frame set is formed by integrating the representative frames extracted by the difference computation from sub-blocks which are at the same position of a frame in the video frame sequence. The key frame set is up for using in the frame interpolation step to interpolate frames between two adjacent key frames in the set with the 3-D DCT algorithm. The experimental results show the advantage on acceleration of the execution speed for performing the fast 2-D DCT and IDCT algorithm that can be more than 7 times compared with the conventional DCT method. In addition, the merit of the block based key frame selection mechanism to detect the variation caused by a small portion of the visual content is presented as well.

[1] N. Dimitrova, H. Zhang, B. Shahraray, I. Sezan, T. Huang, and A. Zakhor, “Applications of video-content analysis and retrieval,” IEEE MultiMedia, vol. 9, no. 3, pp. 44-55, 2002.
[2] P. Geetha and V. Narayanan, “A survey of content-based video retrieval,” Journal of Computer Science, vol. 4, no. 6, pp. 474-486, 2008.
[3] H. Zhang, J. Y. A. Wang, and Y. Altunbasak, “Content-based retrieval and compression: A unified solution,” in Proc. ICIP Santa Barbara, CA, vol. 1, pp. 13-16, 1997.
[4] A. Hanjalic, R. L. Lagendijk, and J. Biemond, “Automated high-level movie segmentation for advanced video retrieval systems,” IEEE Trans. Circuits Syst. Video Technol. , vol. 9, pp. 580-588, 1999.
[5] M. M. Yeung and B. Liu, “Efficient matching and clustering of video shots,” in Proc. IEEE ICIP, vol. 1, pp. 338-341, 1995.
[6] M. Irani and P. Anandan, “Video indexing based on mosaic representations,” Proceeding of the IEEE, vol. 86, no. 5, pp. 905-921, 1998.
[7] A. Hanjalic and H. Zhang, “An integrated scheme for auto-mated video Circus and abstraction based on unsupervised cluster-validity analysis,” IEEE Transactions on Systems for Video Technology, vol. 9, no. 8, pp. 1280-1289, 1999.
[8] T. Liu, H. Zhang, and F. Qi, “A novel video key-frame-extraction algorithm based on perceived motion energy model,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 13, no. 10, pp. 1006-1013, 2003.
[9] J. Jun, S. Lee, Z. He, M. Lee, and E. S. Jang, “Adaptive Key Frame Selection for Efficient Video Coding,” Pacific Rim Symposium on Image Video and Technology, vol. 4872, pp. 853-866, 2007.
[10] A. K. Jain, Fundamentals of Digital Image Processing, Prentice-Hall, Englewood
Cliffs, NJ, 1989.
[11] E. Maeland, “On the comparison of interpolation methods,” IEEE Trans. Medical Imaging, vol. 7, no. 3, pp. 213-217, 1988.
[12] H. S. Hou and H. C. Andrews, “Cubic splines for image interpolation and digital filtering,” IEEE Transactions on Acoustics, Speech and Signal Processing, vol. assp- 26, no. 6, pp. 508-517, 1978.
[13] R. G. Keys, “Cubic convolution interpolation for digital image processing,” IEEE Transactions on Acoustics, Speech and Signal Processing, vol. assp-29, no. 6, pp. 1153-1160, 1981.
[14] M. Unser, A. Aldroubi, and M. Eden, “Fast B-spline transforms for continuous image representation and interpolation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 3, pp. 277-285, 1991.
[15] T. C. Chen and R. J. P. deFigueiredo, “Two-dimensional interpolation by generalized spline filters based on partial differential equation image models,” IEEE Transactions on Acoustics, Speech and Signal Processing, vol. assp-33, no. 3, pp. 631-642, 1985.
[16] K. Jensen and D. Anastassion, “Subpixel edge localization and the interpolation of still images,” IEEE Transactions on Image Processing, vol. 4, no. 3, pp. 285-295, 1995.
[17] S. M. Guo, C. B. Li, C. W. Chen, Y. C. Liao, and S. H. Tsai, “Enlargement and reduction of image/video via discrete cosine transform pair, part 1: novel three-dimensional discrete cosine transform and enlargement,” Journal of Electron, Imaging, vol. 16, no. 4, 043006, pp. 1-19, 2007.
[18] S. M. Guo, C. B. Li, C. W. Chen, Y. C. Liao, and S. H. Tsai, “Enlargement and reduction of image/video via discrete cosine transform pair, part 2: reduction,” Journal of Electron, Imaging, vol. 16, no. 4, 043007, pp. 1-9, 2007.
[19] J. A. Roese, W. K. Pratt, and G. S. Robinson, “Interframe cosine transform image coding,” IEEE Transactions on Communication, vol. 25, no. 11, pp. 1329-1338, 1977.
[20] G. Karlsson and M. Vetterli, “Three dimensional subband coding of video,” in Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, pp. 1100-1103, 1988.
[21] C. I. Podilchunk, N. S. Jayant, and N. Farvardin, “Three-dimensional subband coding of video,” IEEE Transactions on Image Processing, vol. 4, no. 2, pp. 125-139, 1995.
[22] Y. Chan and W. Siu, “A new adaptive interframe transform coding using directional classification,” in Proc. IEEE Int. Conf. Image Processing, vol. 2, pp. 977- 981, 1994.
[23] X. Ji, C. Zhang, J. Y. Wang, and S. H. Boey, “Fast 2-D 8×8 discrete cosine transform algorithm for image coding,” Science in China Series F: Information Sciences, vol. 52, no. 2, pp. 215-225, 2009.
[24] X. Ji and C. Zhang, “A fast IDCT algorithm for image decompression,” Chinese J Comput. , vol. 28, no. 12, pp. 2079-2088, 2005.