Abstract
Two-dimensional image motion is the projection of the three-dimensional motion of objects, relative to a visual sensor, onto its image plane. Sequences of time-orderedimages allow the estimation of projected two-dimensional image motion as either instantaneous image velocities or discrete image displacements. These are usually called the optical flow field or the image velocity field. Provided that optical flow is a reliable approximation to two-dimensional image motion, it may then be used to recover the three-dimensional motion of the visual sensor (to within a scale factor) and the three-dimensional surface structure (shape or relative depth) through assumptions concerning the structure of the optical flow field, the three-dimensional environment, and the motion of the sensor. Optical flow may also be used to perform motion detection, object segmentation, time-to-collision and focus of expansion calculations, motion compensated encoding, and stereo disparity measurement. We investigate the computation of optical flow in this survey: widely known methods for estimating optical flow are classified and examined by scrutinizing the hypothesis and assumptions they use. The survey concludes with a discussion of current research issues.
- ADELSON, E H. AND BERGEN, J R. 1986. The early detection of motion boundaries. In iEEE Proceedings o/' Workshop on Vtsual Motzon (Charleston, S.C., May). 151 156.Google Scholar
- ADELSON, E H. mNo BERGEN, J. R. 1985 Spatiotemporal energy models for the perception of motion d. Opt Soc Am A2, 2,284 299Google Scholar
- ADR5 G. 1985. Determining three-d~mensmnal motion and structure from optical flow generated by several movmg objects. IEEE PAMI 7, 4, 384 401.Google Scholar
- AGGARWAL, J. K AND NANDHAKUMAR, N 1988 On the computation of motion from sequences of images a review. Prom IEEE 76, S, 917 935Google Scholar
- AISBETT, J. 1989. Optical flow w~th intensityweighted smoothing. IEEE PA34I 11, 5 (1984), 512 522. Google Scholar
- ALOIMONOS, J. AND BROWN, C M 1984 Direct processing of curvilinear sensor motion from a sequence of perspective images. In IEEE Proceedzng~ of Workshop on Computer Vlston: Representatmn and Control (Annapolis, Md, April-May), IEEE, 72 77Google Scholar
- ALOiMONOS, J. AND RISTIGOUS, I. 1986 Determining the 3d motion of a m~d planar patch without correspondence, under perspective projection. In IEEE Proceedings of Workshop on Motion, Representatzon and Analyszs (Annapohs, Md., April May), IEEE. 167 174Google Scholar
- ANAND~X5 P. 1989. A computational framework and an algorithm for the measurement of visual motion. Int. J. Comput. -Vzszon 2,283-310.Google Scholar
- ANCONA, N. 1992. A fast obstacle detection method based on optical flow In Proceedings of ECCV (Santa Margherita, Llgure, Italy, May), Springer Verlag, 267-271. Google Scholar
- BARMAN, H. 1991. Hmrarchmal curvature estimation m computer vision, Univ. of Linkopmg, Dept. of Electrical Engineering, Ph.D. Thesis.Google Scholar
- BARMAN, H., HAGLUND, L, KNUTSSON, H., AND GRANLUND, G.H. 1991. Estimatmn of velocity, acceleratmn, and disparity m t~me sequences. In IEEE Proceedings of Workshop on Vzsual Motmn (Irvine, Calif., March), 44 51Google Scholar
- BARNARD, S. T. mwD THOMPSON, W. B 1980 Disparity analysis of images IEEE PAMI 2, 4, 333-340.Google Scholar
- BARREN, J. L. AND EAGLESON, R. 1995. Recurs~ve eshmatmn of time-varying motmn and structural parameters. Pattern Recogn. (m press). Google Scholar
- BARREN, J. L, FLEET, D. J., AND BEAUCHEMIN, S. S. 1994 Performance of optical flow techmques. Int. d. Comput Vis'mn 12, 1, 43 77. Google Scholar
- BARREN, J. L, JEPSON, A D., AND TSeTSeS, J K. 1990 The feaslMhty of motmn and structure from nomy brae-varying image velomty refermatron. Int. J. Comput. Vision 5, 3, 239 269 Google Scholar
- BARREN, J. L. AND LIPTAY, A 1994. Optic flow to measure mmute increments in plant growth. Bmlrnagmg 2, 57-61Google Scholar
- BATrlTi, R, AMALDL E., AND KOCH, C. 1991. Computing optical flow across multiple scales: An adaptive coarse-to-fine strategy. Int. J. Cornput. Vzsmn 6, 2, 133 145. Google Scholar
- BEAUCHEMIN, S S AND BARRON, J. L 1995 The structure of occlusion in Foumer space In Vls/on Interface (Quebec City, Canada, May) 112-119Google Scholar
- BERGEN, J. R., ANANDAN, P., HANNA, K. J., AND E{INGORANi, R. 1992 Hierarchical modelbased motmn estimatmn. In Procee&ngs of ECCV (Santa Marghemta, Italy). Springer Vetlag, 237-252. Google Scholar
- BERGEN, J. R. BURT, P J, HINGORANI, R., AND PELF~G, S. 1992 Three-frame algorithm for estimating two-component image motmn. IEEE PAMI 14, 9, 886-896 Google Scholar
- BLACK, M. J. 1992. Robust incremental optical flow. Yale Umverslty, New Haven, CT, Ph.D. Thesis. Google Scholar
- BLACK, M J. 1991. A robust gradient-method for determining optical flow Tech. Rep. YALEU/DCS/RR-891, Yale University, New Haven, CT.Google Scholar
- BLACK, M. J. AND ANANDAN, P. 1993 A framework for robust estimation of optical flow. In Proceedings oflCCV (Berlin, May), 231 236Google Scholar
- BLACK, M J. ANn ANANDAN, P. 1991 Robust dynamic motion estimation over t~me In Proceedrags IEEE CVPR (Los Alamitos, CAI, 296-302.Google Scholar
- BLACK, M J. AND ANANDAN, P. 1990. A model for the detectmn of motmn over t~me. In Proceedrags of ICCV (Osaka, Dec ), 33 37.Google Scholar
- BLACK, M. J. AND JEPSON, A. 1994 Estimating optical flow in segmented nnages using variable-order parametric models with local deforraations. Tech. Rep SPL-94-053, Xerox Systems and Practices Laboratory, Palo Alto, CAGoogle Scholar
- BOBER, M. AND 14ATTLER, J 1994 Robust morton analyms In IEEE Proceedings of CVPR (Seattle, WA, June), 947-952Google Scholar
- BOUTHEMY, P.~ND FRANCOIS, E. 1993. Motion segmentation and qualitative dynamic scene analyas from an image sequence. Int. J. Cornput Vlston 10, 2, 159-182 Google Scholar
- BURLINA, P. AND CHELLAPPA, R. 1994. Time-to-x: Analysis of motion through temporal parameters. In Proceedings IEEE CVPR (Seattle, WA, June), 461-468Google Scholar
- BURT, P. J. AND ADELSON, E.H. 1983. The Laplaclan pyramid as a compact image code. iEEE Trans. Cornmun. 31,532 540.Google Scholar
- BURT, P. J., HINGORANI, R., AND KOLCZYNSKI, R. J. 1991. Mechanisms for Isolating component patterns in the sequential analysis of multiple motion. In IEEE Proceedings of Workshop on V~sual Morton (Princeton, NJ, Oct.), 187 194.Google Scholar
- BUXTON, B. F. AND BUXTON, H. 1984. Computation of optic flow from the motion of edge features in image sequences. Image Vision Cornpitt 2, 2, 59 75Google Scholar
- CAMPANI, M. AND VERm, A 1990. Computing optical flow from an overconstrained system of linear equations. In Proceedings of ICCV (Osaka, Dec.), 22 26.Google Scholar
- CARPENTIERi, B. AND STORER, J.A. 1992. A splitmerge parallel block-matching algorithm for video displacement estimation. In Data Comprcssion Conference (Snowbird, UT, March), 239-248.Google Scholar
- CmN, T. M., KARL, W., AND WiLLSKY, A. 1994. Probabihstic and sequential computation of optical flow using temporal coherence. IEEE Trans. Image Process. 773 788.Google Scholar
- CHU, C. H. AND DELP, E. J. 1989. Estimating displacement vectors from an image sequence. J. Opt. Soc. Am. A 6, 6, 871-878Google Scholar
- CORNELWS, N. AND KANADE, T. 1983. Adapting optical flow to measure object motion in reflectance and x-ray image sequences. In Procee&ngs of ACM S~ggroph/Szgart Inter&scplinary Workshop on Motion (Toronto, April), ACM, New York, 50-58. Google Scholar
- CORNILLEAU-PERES, Y. AND DROULEZ, J. 1990. Stereo correspondence from optical flow. In Proceedings of ECCV (Antibes, France, April), 326 330. Google Scholar
- DARRELL, T. AND PENTLAND, A. 1991. Robust estimation of a multi-layered motion representation. In IEEE Proceedings of Workshop on Vzsual Motion (Princeton, NJ, Oct.), 173-178.Google Scholar
- DUBOIS, E. 1985. The sampling and reconstruction of time-varying imagery with application in video systems. Proc. IEEE 73, 4. 502 522.Google Scholar
- DUNCAN, J. H. AND CHOU, T. 1992. On the detectmn of motion and the computation of optical flow. IEEE PAMI 14, 3, 346 352. Google Scholar
- DUTTA, R., MANMATHA, R., WILLIAMS, L., AND RISEMAN, E.M. 1989. A data set for quantitative motion analysis. In IEEE Proceedings of CVPR (San Diego, CA, June), 159 164.Google Scholar
- ENKELMANN, W. 1990. Obstacle detection by evaluation of optical flow fields fi'om image sequences. In Proceedings of ECCV (Antibes, France, April). 134-138. Google Scholar
- ENKELMANN, W. 1986. Investigations of multigrid algorithms for the estimation of optical flow fields in image sequences. In IEEE Proceedmgs of Workshop on Motmn: Representation and Analysis (Charleston, SC, May), 81 87.Google Scholar
- FERMIN, I. ANn IMIYA, A. 1994. Two-dimensional motion computation by randomized method. Tech. Rep. TR ICS-4-6-1994, Dept of Information and Computer Sciences, Chiba University, Japan.Google Scholar
- FLEET, D.J. 1992. Measurement of Image Veloc- Ity. Kluwer Academic Publishers, Norwell, MA. Google Scholar
- FLEET, D. J. AND JEPSON, A.D. 1990. Computation of component image velocity from local phase information. Int. J. Comput. Vision 5, 1, 77-104. Google Scholar
- FLEET, D. J. AND LANGLEY, K. 1995a. Computational analysis of non-Fourier motion Vision Res. 34, 22, 3057 3079.Google Scholar
- FLEET, D. J. AND LANGLEY, K. 1995b. Recursive filters for optical flow. IEEE PAMI 17, 1. Google Scholar
- GEMAN, S. AND GEMAN, D. 1984. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE PAMI 6, 6 (Nov.), 721 741.Google Scholar
- GIACttETTI, A., CAMPANI, M., AND TORRE, V. 1994. The use of optical flow for autonomous navigation. In Proceedings of ECCV (Stockholm, May), 146-151. Google Scholar
- GLAZER, F.C. 1987a. Computation of optical flow by multilevel relaxation. Tech. Rep. COINS- TR-87-64, University of Massachusetts.Google Scholar
- GLAZER, F. C. 1987b. Hierarchical gradientbased motion detection. In DARPA Proceedings of Image Understanding Workshop (Los Angeles, CA, Feb.). 733 748.Google Scholar
- GONG, S. AND BRADY, M. 1990. Parallel computation of optic flow. In Proceedings ofECCV (Antibes, France, April), 124-133. Google Scholar
- GRZYWACZ, N M. AND YmLLE, A.L. 1990 A model for the estimate of local velocity by cells in the visual cortex. Proc. Royal Soczety London, B 239, 129 161.Google Scholar
- GUPTA, N., RAC~HAVAN, S., AND KANAL, L. 1993. Robust recovery of image motion. In Proceedrags of Asza Conference on Computer Vzs~on (Osaka, Japan, Nov.).Google Scholar
- HADDADI, N. AND KUO, C.-C. J. 1992. Computation of dense optical flow with a parametric smoothness model. Tech. Rep. USC-SIPI Report 223, Dept. of Electrical Engineering Systems, University of Southern Californm, 1992.Google Scholar
- HAGLUND, L. 1992. Adaptive multidimensional filtering. University of Linkoping, Dept. of Electrical Engineering, Ph.D. ThesisGoogle Scholar
- HANIISCEIACK, P. AND KLETTE, R. 1995. Evaluation of differential methods for image velocity measurement. Comput. Art~f Intell (to appear).Google Scholar
- HANNA, K.J. 1991. Direct multi-resolution estimation of ego-motion and structure from motion In IEEE Workshop on ~%ual Motion (Princeton, NJ), 156 162.Google Scholar
- H,~NA, K. J ~m'D OKAMOTO, N.E. 1993 Combining stereo and motion for direct estimation of scene structure In Proceedings of ICCV (Berlin, May), 357 365.Google Scholar
- H4RALICK, R. M AND LEE, J. S 1982 The facet approach to optic flow In DARPA Proceedings o/Image Understanding Workshop, (Pale Alto, CA, Sept.). 84-93.Google Scholar
- HAY, J.C. 1966 Optical motions and space perception: An extension of Gibson's analysis Psychological Rev 7,3, 6, 550-565.Google Scholar
- HEEGER, D J. 1988. Optmal flow using spatiotemporal filters. Int. d Comput. Vzamn 1, 279 302.Google Scholar
- HEEGER, D J. AND JEPSON, A. D 1992. Subspace methods for recovering rigad matron 2: Algorithm and implementation lnt. J Comput. V~- szon 7, 2, 95 117. Google Scholar
- HEEL, J. 1990 Direct estimation of structure and motion f~:om multiple frames. Tech Rep. 1190, MIT AI Meno, MIT, Cambridge, MA Google Scholar
- HILDRETH, E C 1984 The computatmn of the velomty field. Prec. Royal SoczeO' of London, B 221, 189 220.Google Scholar
- HORN, B. K P 1987 Motion fields are hardly ever ambiguous Int. J. Comput V~smn J, 259-274.Google Scholar
- HORN, B K. P. AND SCt~UNCK, B. G 1981. Determining optmal flow Artzf. Intell. 17, 185-204Google Scholar
- HORN, B. K. P. ANn WELDON, E.J. 1987. Computat(anally efficient methods for recovering translational motion. In Proceedings of ICCV (London, ,June), 2 11Google Scholar
- IRANL M, Rousso, B, AND PELEG, S. 1994. Computing occluding and transparent matrons Int J Comput Vzszon 12, 1, 5-16 Google Scholar
- IRANI, M., Rousso, B, AND PELEG, S. 1994. Recovery of egomotion using image stabilization. In CVPR (Seattle, WA, June), 454 460.Google Scholar
- JAHNE, B. 1990. Motion determinatmn in spacetime images. In Proceedings ofECCV (Antibes, France, June), 161 173 Google Scholar
- JAIN, R. C 1984 Segmentatmn of frame sequences obtained by a moving observer IEEE PAMI 6, 5, 624 629.Google Scholar
- JAIN, R.C. 1983. Direct computation of the focus of expansion IEEE PAMI 5, 1, 58-63.Google Scholar
- JENKIN, M. R M, JEPSON, A. D, AND TSeTSeS, J K. 1991. Techniques for disparity measurement CVGIP 53, 1, 14-30. Google Scholar
- JEPSON, A. D. AND BLACK, M. 1993. Mixture models for optical flow computation. In IEEE Proceedmgs of CVPR (New York, June), 760 761.Google Scholar
- KALIVAS, D S AND SAWCHUK, A.A. 1991 A reguon matching motion estimation algorithm. CVGIP, 54, 2, 275 288 Google Scholar
- KEARNEY, J. K., THOMPSON, W. B., .~ND BOLEY, D L. 1987 Optmal flow estmmtmn: An error analysn~ of gradient-based methods with local optimization. IEEE PAMI 9, 2, 229 244 Google Scholar
- KIRC~NER, H AN~ NmMANN, H. 1992 Finite element method fbr determination of optical flow. Pattern Recogn. 13, 2, 131 141 Google Scholar
- Kocm C., WANG, H. T., BATTm, R, MATmT~, B., AND ZIOMKOWSKI, C. 1991. An adaptive multiscale approach for estimating optical flow Computatmnal theory and physiological implementation. In IEEE Proceedings of Workshop ov Vzsua{ Motion (Princeton, NJ, Oct.), 111 122Google Scholar
- Koch, C, WANG. H. T., MATHER, B., HsU, A., AND SVAREZ, H. 1989. Computmg optical flow in resistive networks and m the primate wsual system. In IEEE Proceedings of Workshop on Vzsual Motion (Irwne, CA, Marcht, 62-72.Google Scholar
- KcRIES, R. AND ZIMMERMAN, G 1986 A versatfie method for the estimation of displacement vector fields from image sequences In IEEE Proceedzngs of Workshop on Motzon Representation and Analysis (Charleston, SC, May), 101 1(16Google Scholar
- LANGLEY, K., ATHERTON, T. J., ~VILSON, R. G., AND LACOIVmE, M. H.E. 1991. Vertical and homzontal disparities from phase. Image Vzswn Comput 9, 4, 296-302.Google Scholar
- LIN, T. AND BARriON, J. L. 1994 Image reconstruction error for optma} flow. In Vision Inter- /ace (Banff National Park, Alberta, May), 713 80.Google Scholar
- LIPTAY, A, BARREN, J. L, JEWETT, T., AND VAN WESENBEECK, l. 1995. Optic flow as an ultra-sensitive technique for measurmg seedling growth in long image sequences. J Am Sac. Hart. Scl. 120, 3, 379-385Google Scholar
- LITTLE, J J. 1992. Accurate early detection of dmcontinmtms. In Vision Interface (Vancouver, B.C, Dec k 2-7Google Scholar
- LITTLE, J J, BUTLHOFF, H H, AND PoC;GrO, T. 1988 Parallel optical flow using local voting In Proceedings of ICCV (Tampa, FL, Dec.), 454-459.Google Scholar
- Lru, H., HERMAN, M , HANG, T.-H., AND CHELLAPPA, R. 1993. A reliable optmal flow algorithm using 3-d Hermite polynommls. Tech. Rep. NIST-IR-5333, National Institute of Standards and Technolog%Google Scholar
- LONGUET-HIGGINS, H. C 1984 The visual ambigmty oi a mowng plane Proc Royal Soczet, y of London, B 223, 165 174.Google Scholar
- LONGUET-HIG~;INS, H C 1981 A computer algorithm for reconstructing a scene from two projections Nature, 223, 133 135Google Scholar
- LONC, UET-HIOGINS, H. C. AND PRAZDNY, K. 1980. The interpretation of a moving retinal nnage. Prec. Royal Socze(~ of London, B 208,385-397.Google Scholar
- Lucns, B. D 1984 Generalized ~mage matching by the method of differences. Carnegie-Mellon Umv, Ph.D. Thesis. Google Scholar
- LUCAS, B. D. AND KANADE, T. 1981. An iterative image-regSstration technique with an application to stereo vision. In Proceedings of IJCAI (Vancouver, B.C., Aug ), 674-679Google Scholar
- LUETTGEN, M. R., KARL, W. C., AND WILLSKY, A. S. 1994. Efficient multiscale reg~llarization with applications to the computation of optical flow. IEEE Trans. Image Process. 3, 1 (Jan.), 41-64.Google Scholar
- MARKANDEY, V AND FLINCHBAUGH, B. E. 1990. Multispectral constraints for optical flow computation. In Procee&ngs of ICCV (Osaka, Japan, Dec.), 38-41.Google Scholar
- MARR, D. AND HILDRETH, E. C. 1980. Theory of edge detection. Proc. Royal Society Londolz, B 290, 187 212.Google Scholar
- DE MICHELI, E., TORRE, V., AND URAS, S. 1993. The accuracy of the computation of optical flow and of the recovery of motion parameters. IEEE PAMI 15, 5, 434-447. Google Scholar
- MITICHE, A., WAN,, Y. F., AND AGGARWAL, J. K. 1987. Experiments in computing optical flow with the gradient-based, multiconstraint method. Pattern Recogn. 20, 2, 173 179. Google Scholar
- MOUNTS, F. W. 1969. A video encoding system using conditional picture-element replenishment. Bell Syst. Tech. J. 48, 2545-2554.Google Scholar
- MUKAWA, N. 1990. Estimation of shape, reflection coefficients and illuminant direction from image sequences. In Proceedzngs of ICCV (Osaka, Japan, Dec.), 507-512.Google Scholar
- MURRAY, D. W. AND BUXTON, B. F. 1987. Scene segmentation from visual motion using global optimization. IEEE PAMI 9, 2, 220-228. Google Scholar
- MURRAY, U. W. AND BUXTON, B.F. 1984. Reconstructing the optic flow from edge motion: An examinatloh of two different approaches. First IEEE Conference on AI Apphcations (Denver, CO, 1984), 382 388.Google Scholar
- MUSMANN, H. G., PIRSCH, P., AND GRALLERT, H. J. 1985 Advances in picture coding. Proc. IEEE, 73, 4, 523-548.Google Scholar
- NAGEL, H.-H. 1990. Extending the 'oriented smoothness constraint' into the temporal domain and the estimation of derivatives of optic flow. In Proceedings ofECCV (Antibes, France, April), 139-148. Google Scholar
- NAGEL, H.-H. 1989. On a constraint equation for the estimation of displacement rates in image sequences. IEEE PAMI 11, 1, 13-30. Google Scholar
- NAGEL, H.-H. 1987. On the estimation of optical flow: Relations between different approaches and some new results. Artif. Intell. 33, 299-324. Google Scholar
- NAGEL, H.-H. 1983a. Constraints for the estimation of vector fields from image sequences. In Proceedzngs IJCAI, (Karlsruhe, Germany, Aug.). 945-951.Google Scholar
- NAaEL, H.-H. 1983b. Displacement vectors derived from second-order intensity variations in image sequences. CVGIP, 21, 85-117.Google Scholar
- NAGEL, H.-H. AND ENKELMANN, W. 1986. An investigation of smoothness constraints for the estimation of displacement vector fields from image sequences. IEEE PAMI 8, 5, 565-593. Google Scholar
- NEGAHDARIPOUR, S. AND HORN, B. K.P. 1987. Direct passive navigation. IEEE PAMI 9, l, 168-176. Google Scholar
- NEGAHDARIPOUR, S. AND LEE, S. 1992. Motion recovery from image sequences using only first order optical flow information. Int. J. Comput. Vision 9, 3, 163-184. Google Scholar
- NEGAHDARIPOUR, S. AND YU, C.H. 1993. A generalized brightness change model for computing optical flow. In Proceedings ICCV (Berlin, May), 2-11.Google Scholar
- NETRAVALI, A. N AND ROBBINS, J.D. 1979. Motion compensated television coding: Part 1. Bell Syst. Tech. J 58, 631 670.Google Scholar
- OGATA, M. AND SATO, T. 1992. Motmn-detection model with two stages: Spatiotemporal filtering and feature matching. J. Opt. Soc. Am. A 9, 3, 377 387.Google Scholar
- OKUTOMI, M. AND KANADE, T 1990. A locally adaptive window for signal matching'. In ProeeedD~gs of ICCV (Osaka, Japan, Dec.), 190 199Google Scholar
- OVERiNGTON, I 1987 Gradient-based flow segmentation and location of the focus of expansion. In Alvey V~sion Conference (Cambridge Univ., Sept.), 860-870.Google Scholar
- PERRONE, J.A. 1990. Simple techmque for optical flow estimation. J. Opt. Soe. Am. A 7 2, 264 277.Google Scholar
- POGGIO, T., GAMBLE, E., AND LITTLE, J. 1988. Parallel integration of vision modules. Science 242, 436-440.Google Scholar
- PRAZDN~, K. 1985. Detection of binocular disparities. Biol. Cyber,. 52, 93 99Google Scholar
- PP~ZDNY, K. 1979. Motion and structure from optical flow. In Proceedings of IJCAI (Tokyo, Aug.), 702-704.Google Scholar
- PRINCE, J. L. AND MCVEIGH, E.R. 1992. Motion estimation from tagged mr image sequences. IEEE Trans. Medical Images 11, 2,238 249Google Scholar
- REGAN, D. AND BEVERLEY, K.I. 1982. How do we avoid confounding the direction we are looking and the direction we are moving. Science 215, 194 196.Google Scholar
- REICHARDT, W., SCHLOGL, R. W., AND EGELHOAF, M. 1988. Movement detectors of the correlation type provide sufficient information for local computation of 2d velocity fields. Naturwissenschaften 75, 313-315.Google Scholar
- RO~NONE, A., CAMPANI, M., AND VERm, A. 1992. Identifying multiple motions from optical flow. In Proceedings of ECCV (Santa Margherita Ligure, Italy, May), 258 266. Google Scholar
- SCHNORR, C. 1992. Computatmn of discontinuous optical flow by domain decomposition. IEEE PAMI 8, 2, 153-165. Google Scholar
- SCHNORR, C. 1991 Determining optical flow for irregular domains by minimizing quadratic functionals of a certain class Int. d Conzput. Vision 6, 1, 25 38. Google Scholar
- SCHUNCK, B. G 1989 Image flow segmentation and estimation by constraint line clustermg. IEEE PAMI 11, 10, 1010 1027 Google Scholar
- SCHUNOK, B.G. 1985. Image flow Fundamentals and future research. In IEEE Proceedings of CVPR, (San Francmco, June), 560 571Google Scholar
- SCOTT, G. L. 1987. Four-line method of locally estimating optic flow Image Vzsmn Comput. 5, 2, 67-72 Google Scholar
- SHfZAWA, M. AND MASE, K 1991 Principle of superposltion: A common computational framework for analysis of multiple motion In IEEE Proceedmgs of Workshop on Visual Motion IPrinceton, NJ., Oct.), 164 172.Google Scholar
- SiMONCELLI, E P., ADELSON, E. H., AND HEEGER, D. J 1991 Probability distributions of optical flow. In IEEE Proceedings of CVPR (Los Alamitos, CA, June), 310 315Google Scholar
- SINGII, A. 1992. Optm flow computation. A urnfled perspectme IEEE Computer Society Press, Los Alamitos, CA.Google Scholar
- SIN~;H, A. 1991 Incremental estimation of image flow using a Kalman filter In IEEE Proceedrags of Workshop on Visual Motion iPrinceton, NJ, Oct.), 36 43.Google Scholar
- SINGH, A. 1990. An estimation-theoretic fkamework Ibr image flow computation. In Proceedrags oflCCV (Osaka, Dec.), 168-177.Google Scholar
- SOBEY, P AND SmNIVASAN, M.V. 1991. Measurement of optical flow by a generalized gradient scheme. J Opt. Soc. Am. A 8, 9, 1488-1498Google Scholar
- SPACEK, L.A. 1986 Edge detection and motion detection, image Vlsmn Comput. 4, 1, 43-56. Google Scholar
- SPETSAKIS, M. E. 1994. An optical flow estimation algorithm that uses Gabor filters and affine model for flow. Tech. Rep., Dept of Computer Science, York University.Google Scholar
- SRINIVASAN, M. V. 1990. Generalized gra&ent schemes for the measurement of two-dimensional image motion. B~ol. Cybern. 6'3,421 431. Google Scholar
- SUBBARAO~ M. 1990. Bounds on time-to-collision and rotational component from first-order derivatives of image flow CVGIP 50, 329-341 Google Scholar
- SUBBARAO, M. AND WAXMAN, A.M. 1985 On the umqueness of image flow solutions for planar surfaces in motion. In Third Workshop on Computer Vision: Representation and Control (Bellaire, MI, Oct ), 129-140.Google Scholar
- SUNDaRESWARaN, V. 1992. A fast method to estimate sensor translation. In Proceedmgs of ECCV (Santa Margherita Ligura, Italy, May), 257 263. Google Scholar
- SUTTON, M. A, WALTERS, W. J., PETERS, W H., RANSON, W F., AND McNEIL, S R. 1983. Determination of disp}acement using an improved digital correlatmn method. Image Vzsmn Cornput 1, 3, 133-139.Google Scholar
- TaOMPSON, W B, MUTCH, K M., .~ND BERZINS, V. A. 1985 Dynaniic occlusion analysis in optical flow fields IEEE PAMI 7, 4, 374 383.Google Scholar
- THOMPSON, W B AND PONG, T. 1990. Detecting moving objects Int d. Comput. Vision 4, 1, 39 57. Google Scholar
- TISTARELLL M. AND SANDINI, G. 1990. Estimatmn of' depth i~om motmn using anthropomorphic visual sensor. Image Vzsmn Comput 8, 271 278. Google Scholar
- TRETIAK, O. AND PASTOR, L 1984. Velocity estimatron from image sequences with second-order differential operators. In IEEE Proceedings of ICPR IMontreal, Quebec, July Oct.). 20-22.Google Scholar
- Ts^i, R. Y. m'4a HUANG, T. S. 1984. Uniqueness and estimation of three-dimensional motion parameters of rigid objects with curved surfaces. IEEE PAMI 6, 1, 13 27.Google Scholar
- TSAI, R. Y., HUANG, T. S., aND ZHU, W 1982. Eshmating three-dimensional motion parameters of a rigid planar patch 2. Singular value decomposition. IEEE Trans. Acoustics, Speech Signal Process 30, 4, 525-534.Google Scholar
- ULLMAN, S. 1979. The Interpretatmn of Visual Motion MIT Press, Cambridge, MA.Google Scholar
- URAS, S, Gmosh F, VERRh A., AND TORRE, V 1988 A computational approach to motion perception Biol Cybern 60, 79 87.Google Scholar
- WANG, J. Y. A AND ADELSON, E H. 1994. Representing moving images with layers IEEE Trans. Image Process 3, 5, 625-638.Google Scholar
- WATSON, A B ANn AHUMADA, A. J JR. 1985 Model of human visual motion sensing- J. Opt Soc. Am. A 2, 2, 322-341.Google Scholar
- WAXMAN, A. M. AND WOHN, a. 1985 Contour evolution, nmghborhood deformation, and global image flow: planar surfaces in motion Int. J. Robotms Res. 4, 3, 95-108.Google Scholar
- WAXMAN, A M., Wu, J, ANn BERGHOLM, F. 1988 Convected activation profiles and the measurement of visual motmn. In IEEE Proceedings of CVPR (Ann Arbor, MI, June), 717 723.Google Scholar
- WEBER, J. AND MALIK, J. 1993. Robust computation of optical flow m a multi-scale differentml framework In Proceedmgs of ICCV (Berlin, May), 12-20.Google Scholar
- WENG, J 1990 A theory of image matching. In Proceedings of ICCV (Osaka, Dec.), 200-209.Google Scholar
- WmTTEN, G 1990. A framework for adaptive scale space tracking solutions to prob}ems in computer vision. In Proceedings of ICCV /Osaka, Dec.), 210-220Google Scholar
- WILLICK, D. AND YANG, Y. 1989. Experimental evaluation of motion constraint equatmns. Tech. Rep. 89-4, Dept. of Computational Science, Univ. of Saskatchewan.Google Scholar
- WoomtmxL R.J. 1990. Multiple light source optical flow. In Proceedings oflCCV (Osaka, Dec.), 42 46.Google Scholar
- X~ON% Y. AND SHAFER, S.A. 1994. Moment and hypergeometric filters for high precision computation of focus, stereo and optical flow. Tech. Rep CMU-RI-TR-94-28. Dept. of Computer Science, Carnegie-Mellon Univ.Google Scholar
- ZHANG, Z. AND FAUGERAS~ O.D. 1992. Three-dimensional motion computation and object segmentation in a long sequence in stereo frames. Int. d. Comput. Vzslon 7, 3. Google Scholar
- ZHENC~, H. AND BLOSTEIN, S.D. 1993. An errorweighted regularization algorithm for image motion-field estimation. IEEE Trans. Image Process. 2, 2, 246 252.Google Scholar
- ZHENG, Q. AND CHELLAPPA, R. 1993. Automatm feature point extraction and tracking in image sequences for unknown image motion. In Proceedmgs oflCCV (Berlin, May), 335-339.Google Scholar
- ZINNER, H. 1986. Determining the kinematic parameters of a moving imaging sensor by processing spatial and temporal intensity changes. J. Opt. Soc. Am. A 3, 9, 1512-1517.Google Scholar
- ZOLTOWSKt, M.D. 1987 Signal processing applications of the method of total least squares. In IEEE 21st Annual Aszlomar Conference on Signals, Systems, and Computers (Pacific Grove, CA, Nov.), 290 296.Google Scholar
Index Terms
- The computation of optical flow
Recommendations
The Accuracy of the Computation of Optical Flow and of the Recovery of Motion Parameters
The accuracy and the dependence on parameters of a general scheme for the analysis of time-varying image sequences are discussed. The approach is able to produce vector fields from which it is possible to recover 3-D motion parameters such as time-to-...
Motion Field and Optical Flow: Qualitative Properties
It is shown that the motion field the 2-D vector field which is the perspective projection on the image plane of the 3-D velocity field of a moving scene, and the optical flow, defined as the estimate of the motion field which can be derived from the ...
Hierarchical properties of multi-resolution optical flow computation
ECCV'12: Proceedings of the 12th international conference on Computer Vision - Volume 2Most of the methods to compute optical flows are variational-technique-based methods, which assume that image functions have spatiotemporal continuities and appearance motions are small. In the viewpoint of the discrete errors of spatial- and time-...
Comments