Temporal registration of video sequences

In this paper, we propose a temporal registration algorithm for video sequences. This algorithm is developed based on a frame-level model of the temporal misalignment often introduced by video processing algorithms, such as compression, frame rate conversion or by video capturing. With this model, the temporal registration is formulated as a constrained minimization of a matching cost, and it is solved using dynamic programming. The proposed temporal registration can also be extended to integrate both spatial registration and histogram registration. In addition, prior information about the application can be easily incorporated in the form of contextual cost. Experimental results on both synthetic and real processed video sequences with temporal misalignments have shown that the proposed algorithm is effective and robust to a wide range of noise.