System and method for detecting and tracking a curvilinear object in a three-dimensional space

1. A system for detecting and tracking a curvilinear object in a three-dimensional space, comprising:

an image acquisition system comprising a video camera arranged to acquire a video image of said curvilinear object and output a corresponding video signal, said video image comprising a plurality n of image frames each at a respective time t i , where i=1, 2, . . . , n; and

a data processing system adapted to communicate with said image acquisition system to receive said video signal,

wherein said data processing system is configured to determine a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding frame of said plurality of image frames of said video image matches a curvilinear image in said frame to a predetermined accuracy to thereby detect and track said curvilinear object from time t 1 to time t n .

2. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 1 , wherein said computational model is a non-uniform rational B-spline model.

3. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 1 , wherein said image acquisition system a depth camera.

4. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 1 , wherein said curvilinear object comprises a texture that is visible in said video image and said data processing system is further configured to determine said texture of said curvilinear object in said three-dimensional space at each time t i by including texture in said computational model of said curvilinear object at each time t i .

5. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 4 , wherein said texture is a color pattern such that some portions of said curvilinear object are distinguishable from other portions based on differences in color.

6. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 1 , wherein said curvilinear object has a substantially invariant length and said computational model of said curvilinear object includes a constant-length constraint on the modeled curvilinear object.

7. A system for detecting and tracking a curvilinear object in a three-dimensional space according to claim 1 , wherein said video camera is a stereo video camera arranged to acquire a stereo video image of said curvilinear object and output a corresponding stereo video signal, said stereo video image comprising a plurality n stereo pairs of image frames each at a respective time t i where i=1, 2, . . . , n,

wherein said data processing system is adapted to communicate with said image acquisition system to receive said stereo video signal,

wherein said data processing system is configured to determine a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding stereo pair of frames of said plurality of stereo pairs of image frames of said stereo video image matches a two-dimensional curvilinear image in said stereo pair of frames to a predetermined accuracy to thereby detect and track said curvilinear object from time t 1 to time t n .

8. A method for detecting and tracking a curvilinear object in a three-dimensional space, comprising:

receiving, by a data processing system, a video signal from an image acquisition system that comprises a video camera arranged to acquire a video image of said curvilinear object and output said video signal, said video image comprising a plurality n of image frames each at a respective time t i , where i=1, 2, . . . , n; and

determining a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i using said data processing system such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding frame of said plurality of image frames of said video image matches a curvilinear image in said frame to a predetermined accuracy to thereby detect and track said curvilinear object from time t 1 to time t n .

9. A method for detecting and tracking a curvilinear object in a three-dimensional space according to claim 8 , wherein said computational model is a non-uniform rational B-spline model.

10. A method for detecting and tracking a curvilinear object in a three-dimensional space according to claim 8 , wherein said curvilinear object comprises a texture that is visible in said video image and said data processing system is further configured to determine said texture of said curvilinear object in said three-dimensional space at each time t i by including texture in said computational model of said curvilinear object at each time t i .

11. A method for detecting and tracking a curvilinear object in a three-dimensional space according to claim 10 , wherein said texture is a color pattern such that some portions of said curvilinear object are distinguishable from other portions based on differences in color.

12. A method for detecting and tracking a curvilinear object in a three-dimensional space according to claim 8 , wherein said curvilinear object has a substantially invariant length and said computational model of said curvilinear object includes a constant-length constraint on the modeled curvilinear object.

13. A method for detecting and tracking a curvilinear object in a three-dimensional space according to claim 8 , wherein said video camera is a stereo video camera arranged to acquire a stereo video image of said curvilinear object and output a corresponding stereo video signal, said stereo video image comprising a plurality n stereo pairs of image frames each at a respective time t i where i=1, 2, . . . , n,

wherein said determining a position, orientation and shape of said curvilinear object in said three-dimensional space comprises determining a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding stereo pair of frames of said plurality of stereo pairs of image frames of said stereo video image matches a two-dimensional curvilinear image in said stereo pair of frames to a predetermined accuracy to thereby detect and track said curvilinear object from time t 1 to time t n .

14. A non-transitory computer-readable medium comprising computer-executable code for detecting and tracking a curvilinear object in a three-dimensional space, said computer-executable code comprising instructions that, when executed by said computer, causes said computer to:

receive a video signal from an image acquisition system that comprises a video camera arranged to acquire a video image of said curvilinear object and output said video signal, said video image comprising a plurality n of image frames each at a respective time t i , where i=1, 2, . . . , n; and

determine a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding frame of said plurality of image frames of said video image matches a curvilinear image in said frame to a predetermined accuracy to thereby detect and track said curvilinear object from time t 1 to time t n .

15. The non-transitory computer-readable medium according to claim 14 , wherein said computational model is a non-uniform rational B-spline model.

16. The non-transitory computer-readable medium according to claim 14 , wherein said curvilinear object comprises a texture that is visible in said video image and said data processing system is further configured to determine said texture of said curvilinear object in said three-dimensional space at each time t i by including texture in said computational model of said curvilinear object at each time t i .

17. The non-transitory computer-readable medium according to claim 16 , wherein said texture is a color pattern such that some portions of said curvilinear object are distinguishable from other portions based on differences in color.

18. The non-transitory computer-readable medium according to claim 14 , wherein said curvilinear object has a substantially invariant length and said computational model of said curvilinear object includes a constant-length constraint on the modeled curvilinear object.

19. The non-transitory computer-readable medium according to claim 14 , wherein said video camera is a stereo video camera arranged to acquire a stereo video image of said curvilinear object and output a corresponding stereo video signal, said stereo video image comprising a plurality n stereo pairs of image frames each at a respective time t i , where i=1, 2, . . . , n,

wherein said determining a position, orientation and shape of said curvilinear object in said three-dimensional space comprises determining a position, orientation and shape of said curvilinear object in said three-dimensional space at each time t i by forming a computational model of said curvilinear object at each time t i such that a projection of said computation model of said curvilinear object at each time t i onto a corresponding stereo pair of frames of said plurality of stereo pairs of image frames of said stereo video image matches a two-dimensional curvilinear image in said stereo pair of frames to a predetermined accuracy to thereby detect and track said curvilinear object from time t i to time t n .