## Computation of Optical Flow Using Dynamic Programming (1996)

### Cached

### Download Links

- [clips.imag.fr]
- [b2.cvl.iis.u-tokyo.ac.jp]
- [clips.imag.fr]
- DBLP

### Other Repositories/Bibliography

Venue: | In IAPR Workshop on Machine Vision Applications |

Citations: | 14 - 9 self |

### BibTeX

@INPROCEEDINGS{Quenot96computationof,

author = {Georges M. Quenot},

title = {Computation of Optical Flow Using Dynamic Programming},

booktitle = {In IAPR Workshop on Machine Vision Applications},

year = {1996},

pages = {249--52}

}

### OpenURL

### Abstract

This paper presents an original algorithm for the computation of optical ow called Orthogonal Dynamic Programming (ODP) as well as several enhancements to it. The principle is to minimize a sum of square di erences (SSD) between a pair of images. The originality of the approach is that an optimal matching is searched for entire image strips rather than for pixel neighborhoods. Dynamic programming is used to provide very robust strip alignmentsandamultiresolution iterative process is used to compute the velocity eld. Extensions to the computation of the velocity eldfornoninteger image indexes, to the use of more than two images, and to the search for subpixel velocities, are presented. Results obtained for the Barron, Fleet and Beauchemin performance tests appear to be at least as good as or better than those obtained using classical optical ow detection methods. 1

### Citations

1055 | S.: Performance of optical flow techniques
- BARRON, FLEET, et al.
- 1994
(Show Context)
Citation Context ...n those obtained using classical optical ow detection methods. 1 Introduction Several techniques have been developed for the computation of optical ow. In a survey and a comparative performance study =-=[1]-=-, Barron, Fleet and Beauchemin classify them in four categories: di erential, correlation based, energy based, and phase based. The method proposed here can be considered as a correlation based one si... |

216 | Stereo by intra- and inter-scanline search using dynamic programming
- Ohta, Kanade
- 1985
(Show Context)
Citation Context ... search of monodimensional alignments (the reason is that no natural order can be found for a multidimensional set) and uneasy to use directly for image matching although some attempts have been made =-=[3]-=- [4] [5]. The originality of the Orthogonal Dynamic Programming (ODP) algorithm [2] is that it transforms the search problem for two-dimensional alignments into a carefully selected sequence of search... |

44 |
Dynamic Programming Optimization for Spoken Word Recognition
- Sakoe, Chiba
- 1978
(Show Context)
Citation Context ...iced into parallel and overlapping strips (Figure 1a). Corresponding strips are aligned (Figure 1b) using dynamic programming exactly as 2D representations of speech signal are with the DTW algorithm =-=[6]-=-. An optimal segment to segment alignment is searched for as a continuous and increasing path inside the disparity matrix along which the sum of the local distance (segment to segment L1 or L2 distanc... |

5 |
Pattern Recognition through Dynamic Programming
- Missakian, Zavidovique
- 1985
(Show Context)
Citation Context ...rch of monodimensional alignments (the reason is that no natural order can be found for a multidimensional set) and uneasy to use directly for image matching although some attempts have been made [3] =-=[4]-=- [5]. The originality of the Orthogonal Dynamic Programming (ODP) algorithm [2] is that it transforms the search problem for two-dimensional alignments into a carefully selected sequence of search pro... |

5 | Image matching using dynamic programming: Application to stereovision and image interpolation - Quenot - 1996 |

5 | The \Orthogonal Algorithm" for Optical Flow Detection using Dynamic Programming
- Quenot
- 1992
(Show Context)
Citation Context ...her than for pixel neighborhoods. Dynamic programming is used to provide very robust strip alignments and a multiresolution iterative process is used to compute the ow eld. This method, introduced in =-=[2]-=- for optical ow detection from a pair of images, has been extended to be able to operate on longer sequences of images. It is also able to compute directly the ow eld for Address: LIMSI-CNRS, BP 133, ... |

4 | The "Orthogonal Algorithm" for Optical Flow Detection using Dynamic Programming - Qu'enot - 1992 |

4 |
Dynamic Programming for Region Based Pattern Recognition
- Adam, Burg, et al.
- 1986
(Show Context)
Citation Context ...of monodimensional alignments (the reason is that no natural order can be found for a multidimensional set) and uneasy to use directly for image matching although some attempts have been made [3] [4] =-=[5]-=-. The originality of the Orthogonal Dynamic Programming (ODP) algorithm [2] is that it transforms the search problem for two-dimensional alignments into a carefully selected sequence of search problem... |

3 |
Stereo by intra- and inter-scanline search using dynamic programming
- Otha, Kanade
- 1989
(Show Context)
Citation Context ... search of monodimensional alignments (the reason is that no natural order can be found for a multidimensional set) and uneasy to use directly for image matching although some attempts have been made =-=[3]-=- [4] [5]. The originality of the Orthogonal Dynamic Programming (ODP) algorithm [2] is that it transforms the search problem for two-dimensional alignments into a carefully selected sequence of search... |

1 |
The "Orthogonal Algorithm" for Optical Flow Detection using Dynamic Programming", Pm
- Qudnot
- 1992
(Show Context)
Citation Context ...than for pixel neighborhoods. Dynamic programming is used to provide very robust strip alignments and a multiresolution iterative process is used to compute the flow field. This method, introduced in =-=[2]-=- for optical flow detection from a pair of images, has been extended to be able to operate on longer sequences of images. It is also able to compute directly the flow field for * *ddress: LIMSI-CNRS, ... |

1 | Image Matching using Dynamic Programming: Application to Stereovision and Image Interpolation - Qudnot - 1996 |