Abstract
This paper describes a method for retrieving scenes from a motion picture database by using motion information as a key. The method has three steps: the automatic estimation of motion vectors in frame sequences, the description of motions in spatio-temporal space, and the retrieval of sequences of images. The motion vectors are estimated by block matching. Estimated motion vectors are mapped in spatio-temporal space (x, y, t). Motion vectors in a scene are aggregated into several representative vectors by statistical analysis. The retrieval of scenes is divided into two parts: the specification of query conditions, and matching between the query conditions and the motion database. Similarity is defined in spatio-temporal space as the distance between the query conditions and the stored motion index. Candidate scenes are ranked in order of distance. Experimental results indicate that the proposed method is effective.
Similar content being viewed by others
References
Abe S, Tonomura Y (1992) Scene retrieval method using temporal condition changes (in Japanese). Trans Institute of Electronics, Information and Communication Engineers J75-DII:512–519
Aggarwal JK, Nandhakumar N (1988) On the computation of motion from sequences of images—a review. Proc IEEE 76:917–935
Akutsu A, Tonomura Y, Hashimoto H, Ohba Y (1992) Camera operation extraction from video data (in Japanese). Trans Institute of Electronics, Information and Communication Engineers 75-DII:226–235
Chang NS, Fu KS (1980) Query by pictorial example. IEEE Trans Software Eng 6:519–524
Chang SK, Dimitrof CW, Arndt T (1988) An intelligent image database system. IEEE Trans Software Eng 14:681–688
Cheng C.L, Aggarwal JK (1990) A two-stage hybrid approach to the correspondence problem via forward-searching and backward correcting. Proc 10th International Conference on Pattern Recognition, pp 173–179
Gould K, Shah M(1989) The trajectory primal sketch: a multi scale scheme for representing motion characteristics. Proc 9th International Conference on Pattern Recognition, pp 79–85
Guttman A (1984) R-TREES: a dynamic index structure for spatial searching. ACM Sigmod 14:47–57
Jasinschi RS (1991) The properties of space-time sampling and the extraction of the optical flow: the effects of motion uncertainty. J Visual Commun Image Representation 2:222–229
Jenkin M (1983) Tracking three-dimensional motion light displays. Proc Workshop on Motion: Representation Contr, pp 66–77
Kappagantula S, Rao KR (1985) Motion compensated interframe image prediction. IEEE Trans Commun 33:1011–1015
Musmann HG, Pirsch P, Grallert HJ (1985) Advances in picture coding. Proc IEEE 73:523–548
Ohta Y, Yamada H (1989) Pattern matching using dynamic programming (in Japanese). J Information Processing Society Japan 30:1058–1066
Okuno T, Kume H, Haga T, Yoshizawa T (1971) Multivariate statistical analysis, part I (in Japanese), Nikka-giren, Tokyo
Okuno T, Haga T, Yajima K, Okuno C, Hashimoto S, Furakawa Y (1976) Multivariate statistical analysis, part II (in Japanese), Nikka-giren, Tokyo
Orchard M (1989) A comparison of techniques for estimating block notion in image sequences coding. SPIE Visual Commun Image Processing IV 1199:248–258
Roussopoulos N, Leifker D (1985) Direct spatial search on pictorial databases using packed R-trees. ACM SIGMOD 14:17–31
Sethi IK, Jain R (1987) Finding trajectories of feature points in a monocular image sequence. IEEE Trans Pattern Analysis Machine Intelligence 9:56–73
Srinivasan R, Rao KR (1985) Predictive coding based on efficien motion estimation. IEEE Trans Commun 33:888–896
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Ioka, M., Kurokawa, M. Estimation of motion vectors and their application to scene retrieval. Machine Vis. Apps. 7, 199–208 (1994). https://doi.org/10.1007/BF01211664
Issue Date:
DOI: https://doi.org/10.1007/BF01211664