Motion video retrieval based on optical flow and Haar wavelet

Developments in image processing techniques have made an easy retrieval for digital images. However, applications requiring content-based querying and searching of videos still remain challenging due to their huge amount of data. This paper presents an efficient algorithm based on optical flow to dig out the motion information contained in a given video shot. The corresponding indexical structure is extracted by Haar wavelet. Experimental results show that the method performs well and the extracted index is used to efficiently locate features close to the query point.