隨著科技的發展，立體影片越來越容易拍攝，數量也跟著提升。而影片的穩定化一直是個重要的研究方向。本篇論文提出一個能夠維持影片內容深度相對關係的立體影片穩定方法。
首先，將影片中追蹤到的特徵點，依照影片中出現的物體和其重要程度分群。接著我們對影片中存在的震動分成旋轉和移動兩種，為了去除旋轉方面的震動，會先依靠特徵點估計影片各個畫面的角度相對關係，得到角度相對關係後便可以知道影片大致的旋轉過程，在旋轉的過程中，相近的畫面旋轉的角度也應該是相近的，因此找出那些旋轉角度過大或過小者，減少或增加其旋轉角度，便可以達到和原本的旋轉保持一致的方向，又能夠去除震動且讓畫面轉動得更加流暢，再來，對已經解決旋轉方面震動問題的特徵點，將他們的軌跡大略的分段，因為影片的軌跡是可能改變的，為了不要讓時間點已經距離很遠的軌跡互相影響，故將整段軌跡分成幾個子線段來處理，接下來利用貝茲曲線得到這些子線段穩定化後的子目標路徑，將這些子目標路徑有時間重疊的部分用內插的方式重合，進而得到完整的目標軌跡，這個部分和先前的研究相比，我們提出了一個不同的分段方式，非固定時間長度就切割，這樣的方式會讓我們更接近影片原始的軌跡。最後，利用最佳化的方式，一方面保持畫面內容的形狀，一方面讓各個特徵點盡可能落在對應的目標路徑上，並將求得的影片超出邊界部分裁去。

Filming stereo videos has become easier with the development of science and technology. Thus, there are more and more stereo videos on the internet. Video stabilization has always been an important research topic. This thesis present a method to stabilize stereo videos without change the distance between objects in the frame.
First of all, we need to track for the feature points and separate them into many groups. We suppose the shaky motion not only cause by translation, but also rotation. Thus, smooth the path directly would not similar to the trajectory, which videographer wants. We solve the shakiness of rotation before we smooth the path. To solve the shakiness of rotation, we will estimate angles between two adjacent frames at the beginning. As we know the angles change of all the frames, we can find out the preference of rotation in the video. While a videographer rotate the camera when filming videos consciously, he may hope the angular velocity is a constant. We can find out the inconsistent angular velocity. Via rotate the frame to solve the shakiness of rotation. Then, we use Bézier curve to smooth the trajectories. We split a trajectory into a set of subtrajectories and smooth the subtrajectories independently. Compared with previous research, we split trajectory according to features tracking rate to get the similar trajectories to original video path. After the subtrajectories smoothness, merge them to get the smoothed trajectory. At the joint of two subtrajectories we replace it by interpolation of them. In the end, we execute optimization on smoothness and context preserving to stabilize videos without too much clipping.

[1] Y. Matsushita, E. Ofek, W. Ge, X. Tang and H.-Y. Shum, "Full-frame video stabilization with motion inpainting," IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006.
[2] C. Morimoto and R. Chellappa, "Evaluation of image stabilization algorithms," Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference, 1998.
[3] K.-Y. Lee, Y.-Y. Chuang, B.-Y. Chen and M. Ouhyoung, "Video stabilization using robust feature trajectories," Computer Vision, 2009 IEEE 12th International Conference, 2009.
[4] M. Grundmann, V. Kwatra and I. Essa, "Auto-directed video stabilization with robust L1 optimal camera paths," Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference, 2011.
[5] M. Grundmann, V. Kwatra, D. Castro and I. Essa, "Calibration-free rolling shutter removal," Computational Photography (ICCP), 2012 IEEE International Conference, 2012.
[6] S. Liu, L. Yuan, P. Tan and J.-T. Sun, "Bundled Camera Paths for Video Stabilization".ACM Transactions on Graphics (SIGGRAPH 2013).
[7] H. Guo, S. Liu, T. He, S. Zhu, B. Zeng and M. Gabbouj, "Joint Video Stitching and Stabilization From Moving Cameras," IEEE Transactions on Image Processing, 2016.
[8] Y.-S. Wang, F. Liu, P.-S. Hsu and T.-Y. Lee, "Spatially and Temporally Optimized Video Stabilization," IEEE Transactions on Visualization and Computer Graphics, 2013.
[9] L. Zhang, Q.-K. Xu and H. Huang, "A Global Approach to Fast Video Stabilization," IEEE Transactions on Circuits and Systems for Video Technology, 2 2017.
[10] F. Liu, M. Gleicher, H. Jin and A. Agarwala, "Content-preserving warps for 3D video stabilization," SIGGRAPH '09 ACM SIGGRAPH 2009 papers, 2009.
[11] S. Liu, Y. Wang, L. Yuan, J. Bu, P. Tan and J. Sun, "Video Stabilization with a Depth Camera," Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference, 2012.
[12] B. M. Smith, L. Zhang, H. Jin and A. Agarwala, "Light Field Video Stabilization," IEEE International Conference on Computer Vision (ICCV), 2009.
[13] C. Jia, Z. Sinno and B. L. Evans, "Real-time 3D rotation smoothing for video stabilization".Signals, Systems and Computers, 2014 48th Asilomar Conference.
[14] A. Goldstein and R. Fattal, "Video stabilization using epipolar geometry," ACM Transactions on Graphics (TOG), 2012.
[15] F. Liu, M. Gleicher, J. Wang, H. Jin and A. Agarwala, "Subspace video stabilization," ACM Transactions on Graphics (TOG), 2011.
[16] H. Guo, S. Liu, S. Zhu and B. Zeng, "Joint bundled camera paths for stereoscopic video stabilization," Image Processing (ICIP), 2016 IEEE International Conference, 2016.
[17] M. Grundmann, V. Kwatra, M. Han and I. Essa, "Efficient hierarchical graph-based video segmentation," Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference, 2010.
[18] S. Goferman, L. Zelnik-Manor and A. Tal, "Context-Aware Saliency Detection," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, 10 2012.
[19] C. Tomasi and T. Kanade, "Detection and Tracking of Point Features," Technical Report CMU-CS-91-132, 4 1991.
[20] J. Shi and C. Tomasi, "Good features to track," Computer Vision and Pattern Recognition, 1994. Proceedings CVPR '94., 1994 IEEE Computer Society Conference, 1994.
[21] S.-S. Lin, C.-H. Lin, I.-C. Yeh, S.-H. Chang, C.-K. Yeh and T.-Y. Lee, "Content-Aware Video Retargeting Using Object-Preserving Warping," IEEE Transactions on Visualization and Computer Graphics, 10 2013.