Soft transition analysis technique for video sequences

Video indexing is a technique used where objects within the sequence are segmented, indexed and kept in the database so that they can be retrieved for future reference. One of the problems that arise in the indexing is the occurrence of special effects such as fading and wiping. During the indexing, the frames within such effects are discarded, as they are too complex to be segmented. It is desirable to deduce a technique so that these frames can be utilised in indexing. In this paper, we introduce a method that analyses and decodes the frames within fade transition. This utilises the frame pixel values within the transition to reconstruct the frames in their original sequences, which they are made from. The experiment shows that this technique can successfully reconstruct the original frames even with the existence of limited motion within the objects.

[1]  Warnakulasuriya Anil Chandana Fernando,et al.  Automatic detection of fade-in and fade-out in video sequences , 1999, ISCAS'99. Proceedings of the 1999 IEEE International Symposium on Circuits and Systems VLSI (Cat. No.99CH36349).

[2]  W. J. Heng,et al.  Transition Type-Independent Boundary Refinement Technique for Post Shot Boundary Detection , 1999 .

[3]  Stuart Jay Golin.  New metric to detect wipes and other gradual transitions in video , 1998, Electronic Imaging.

[4]  R. Brunelli,et al.  A Survey on the Automatic Indexing of Video Data, , 1999, J. Vis. Commun. Image Represent..

[5]  Touradj Ebrahimi,et al.  MPEG-4 natural video coding - An overview , 2000, Signal Process. Image Commun..

[6]  Wei Xiong,et al.  Efficient Scene Change Detection and Camera Motion Annotation for Video Classification , 1998, Comput. Vis. Image Underst..