## Good features to track (1994)

### Cached

### Download Links

Venue: | CVPR94 |

Citations: | 1552 - 13 self |

### BibTeX

@MISC{Shi94goodfeatures,

author = {Jianbo Shi and Carlo Tomasi},

title = {Good features to track},

year = {1994}

}

### Years of Citing Articles

### OpenURL

### Abstract

No feature-based vision system can work unless good features can be identified and tracked from frame to frame. Although tracking itself is by and large a solved problem, selecting features that can be tracked well and correspond to physical points in the world is still hard. We propose a feature selection criterion that is optimal by construction because it is based on how the tracker works, and a feature monitoring method that can detect occlusions, disocclusions, and features that do not correspond to points in the world. These methods are based on a new tracking algorithm that extends previous Newton-Raphson style search methods to work under affine image transformations. We test performance with several simulations and experiments.

### Citations

2090 | An iterative image registration technique with an application to stereo vision
- Lucas, Kanade
- 1981
(Show Context)
Citation Context ...ifference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation =-=[10]-=-, [1] and linear image deformation [6], [8], [11], possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard ... |

460 |
A computational framework and an algorithm for the measurement of visual motion
- Anandan
- 1989
(Show Context)
Citation Context ...cognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], [15], [18], [7], [17] and sum-of-squared-difference (SSD) methods [2], =-=[1]-=- show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image defor... |

324 | Computational Frameworks for the Fast Fourier Transform - Loan - 1992 |

132 |
Obstacle Avoidance and Navigation in the Real World by a Seeing Robot Rover
- Moravec
- 1980
(Show Context)
Citation Context ..., [11], possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile =-=[13]-=-, the presence of zero crossings of the Laplacian of the image intensity [12], and corners [9], [5]. Yet, even a region rich in texture can be poor. For instance, it can straddle a depth discontinuity... |

129 |
Gray-level corner detection
- Kitchen, Rosenfeld
- 1982
(Show Context)
Citation Context ...easure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13], the presence of zero crossings of the Laplacian of the image intensity [12], and corners =-=[9]-=-, [5]. Yet, even a region rich in texture can be poor. For instance, it can straddle a depth discontinuity or the boundary of a reflection highlight on a glossy surface. In either case, the window is ... |

80 |
Volumetric model and 3D trajectory of a moving car derived from monocular TV-frame sequence of a street scene
- Dreschler, Nagel
- 1981
(Show Context)
Citation Context ...e of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13], the presence of zero crossings of the Laplacian of the image intensity [12], and corners [9], =-=[5]-=-. Yet, even a region rich in texture can be poor. For instance, it can straddle a depth discontinuity or the boundary of a reflection highlight on a glossy surface. In either case, the window is not a... |

67 |
A locally adaptive window for signal matching
- Okutomi, Kanade
- 1992
(Show Context)
Citation Context ...frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation [6], [8], [11], possibly with adaptive window size=-=[14]-=-. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13], the presence of zero crossings of the La... |

64 |
Algorithms for subpixel registration
- Tian, Huhns
- 1986
(Show Context)
Citation Context ...tion IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], [15], [18], [7], =-=[17]-=- and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with re... |

40 |
Local Correlation Measures for Motion Analysis
- Burt, Yen, et al.
- 1982
(Show Context)
Citation Context ...rn Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], [15], [18], [7], [17] and sum-of-squared-difference (SSD) methods =-=[2]-=-, [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image ... |

29 |
Reliability analysis of parameter estimation in linear models with applications to mensuration problems in computer vision
- FĂ¶rstner
- 1987
(Show Context)
Citation Context ...hat all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation =-=[6]-=-, [8], [11], possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity pro... |

16 |
Bandpass channels, zero-crossings, and early visual information processing
- Marr, Ullman, et al.
- 1979
(Show Context)
Citation Context ...ed based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13], the presence of zero crossings of the Laplacian of the image intensity =-=[12]-=-, and corners [9], [5]. Yet, even a region rich in texture can be poor. For instance, it can straddle a depth discontinuity or the boundary of a reflection highlight on a glossy surface. In either cas... |

15 |
Motion displacement estimation using and affine model for matching
- Fuh, Maragos
- 1991
(Show Context)
Citation Context ...ll the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation [6], =-=[8]-=-, [11], possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile ... |

13 |
Methods for measuring small displacements of television images
- Caorio, Rocca
- 1976
(Show Context)
Citation Context ...xperiments. 1 Introduction IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], =-=[3]-=-, [15], [18], [7], [17] and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some mat... |

10 |
Photogrammetric Standard Methods and Digital Image Matching Techniques for High Precision Surface Measurements
- Foerstner, Pertl
- 1986
(Show Context)
Citation Context ...roduction IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], [15], [18], =-=[7]-=-, [17] and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion w... |

9 |
Prediction of correlation errors in stereo-pair images
- Ryan, Gray, et al.
- 1980
(Show Context)
Citation Context ...ments. 1 Introduction IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], =-=[15]-=-, [18], [7], [17] and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching ... |

8 |
Properties of frame-difference signals generated by moving images
- Connor, Limb
- 1974
(Show Context)
Citation Context ...and experiments. 1 Introduction IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation =-=[4]-=-, [3], [15], [18], [7], [17] and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing som... |

6 |
Realities of automatic correlation problem
- Wood
- 1983
(Show Context)
Citation Context ... 1 Introduction IEEE Conference on Computer Vision and Pattern Recognition (CVPR94) Seattle, June 1994 Is feature tracking a solved problem? The extensive studies of image correlation [4], [3], [15], =-=[18]-=-, [7], [17] and sum-of-squared-difference (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criter... |

4 |
Extracting affine deformations from image patches - i: Finding scale and rotation
- Manmatha, Oliensis
- 1993
(Show Context)
Citation Context ...e basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation [6], [8], =-=[11]-=-, possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13], ... |

2 |
Motion displacement estimation using an a ne model for matching
- Fuh, Maragos
- 1991
(Show Context)
Citation Context ...ll the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation [6], =-=[8]-=-, [11], possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity pro le [... |

2 | Reliability analysis of parameter estimation in linear models with applications to mansuration problems in computer vision - Foerstner - 1987 |

1 |
Properties of framedi erence signals generated by moving images
- Connor, Limb
- 1974
(Show Context)
Citation Context ...to work under a ne image transformations. We test performance with several simulations and experiments. 1 Introduction Is feature tracking a solved problem? The extensive studies of image correlation =-=[4]-=-, [3], [15], [18], [7], [17] and sum-of-squared-di erence (SSD) methods [2], [1] show that all the basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some... |

1 |
Extracting a ne deformations from image patches - I: Finding scale and rotation
- Manmatha, Oliensis
- 1993
(Show Context)
Citation Context ...e basics are in place. With small inter-frame displacements, a window can be tracked by optimizing some matching criterion with respect to translation [10], [1] and linear image deformation [6], [8], =-=[11]-=-, possibly with adaptive window size[14]. Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity pro le [13], t... |

1 | Methods for measuring small displacements in t plevision images - Cafforio, Rocca |

1 | Bandpass channels, zero-crossings, and early visual information processing - Man, Poggio, et al. - 1979 |