以不均勻比例及剪裁技術實作視訊縮放

Video Retargeting Using Non-homogeneous Scaling and Cropping

游文俞(Yo, Wenyu)

指導教授 ： 柳 金 章

國立中正大學/工學院/資訊工程學系/碩士(2012年)

摘要

視訊縮放是用來改變影片的長寬比例並且要盡可能的保持影片中人眼關注物件的原比例，而影像縮放已經有不錯的效果，但在視訊中因為必須要維持時間軸的連貫性，所以仍然是一個難以處理的問題。
本研究利用解方程式的方法來找到縮放數值圖，縮放數值圖表示的是原圖中每個區塊所要縮放的比例，原圖能透過此圖轉換到目標比例的縮放圖。方程式的資訊包括了影片內容物的人眼關注值和影片時間軸的資訊，我們透過此方程式所找的縮放數值圖是可以完滿的保持以上兩者。最後在透過剪裁技術改進結果。本研究之實驗結果顯示，本研究所提方法比其他方法有較好的效能。

關鍵字

不均勻縮放 ； 影像縮放 ； 視訊縮放

並列摘要

Video retargeting is used to change the ratio between height and weight of original size and maintain original size of the saliency areas as possible. Although the efficacy of image retargeting is very good, there is a problem of video retargeting which must maintain the temporal coherence of video. Therefore, video retargeting is an ill-posed problem in image processing.

The proposed method allows solving optimized function to find the scaling factor map. Scaling factor map represents the scaling degree of one direction of each pixel (block) and then we can use the scaling factor map to transform the original video to the retargeting video with target size by per-frame processing. The optimized function includes visual attention value of video content and temporal information. Therefore, the optimized scaling factor map which is solved from optimized function can maintain two above character. Finally, we use the cropping method to improve the efficiency of the optimized scaling factor map. Based on the experimental results obtained in this study, the proposed method has better performance than the previous methods.

並列關鍵字

Non-homogeneous Scaling ； Image retargeting ； Video retargeting

參考文獻

[1] S. Avidan and A. Shamir, “Seam carving for content-aware image resizing,” ACM Trans. on Graphics, vol. 26, no. 3, July 2007.

[2] M. Rubinstein, S. Avidan, and A. Shamir, “Improved seam carving for video retargeting,” ACM Trans. on Graphics, vol. 27, no. 3, Aug. 2008.

[4] C. K. Chiang, S. F. Wang, Y. L. Chen, and S. H. Lai, “Fast JND-based video carving with GPU acceleration for real-time video retargeting,” IEEE Trans. on Circuits and Systems for Video Technology, vol. 19, no. 11, pp. 1588-1597, Nov. 2009.

[5] D. Han, X. Wu, and M. Sonka, “Optimal multiple surfaces searching for video/image resizing – a graph theoretic approach,” in Proc. of 2009 IEEE Int. Conf. on Computer Vision, Sept. 2009, pp. 1026-1033.

[13] Y. S. Wang, C. L. Tai, O. Sorkine, and T. Y. Lee, “Optimized scale-and-stretch for image resizing,” ACM Trans. on Graphics, vol. 27, no. 5, Dec. 2008.