Relative range camera calibration

1. A calibration system for calibrating a first range camera and a second range camera, comprising:

a first data point sampled before a time t using the first range camera;

a second data point sampled after time t using the first range camera;

an interpolation module that interpolates a data point at time t using the first data point and the second data point such that the interpolated data point approximates where a data point would have been seen at time t;

a data matching processor that matches the interpolated data point with data sampled from the second range camera to calibrate the first and the second range cameras;

an error minimization processor that uses synchronized data to compute transformation parameters having a minimum error; and

an output module that expresses data captured by the first range camera and the second range camera in a base coordinate frame using the transformation parameters and without the use of calibration targets or charts.

2. The calibration system of claim 1 , wherein the data is obtained from the range cameras measuring a path of a moving object.

3. The calibration system of claim 1 , wherein the interpolation module uses linear interpolation to obtain the interpolated data point.

4. The calibration system of claim 1 , further comprising a data synchronizer that synchronizes data sampled from the first range camera with data sampled from the second range camera.

5. The calibration system of claim 1 , further comprising a coordinate selector that selects a base coordinate system for use in the calculation of the transformation parameter.

6. The calibration system of claim 4 , wherein the data synchronizer uses a time offset value to synchronize the data and the time offset value is used by the error minimization processor to compute the transformation parameter.

7. A computer-implemented method for calibrating range cameras, comprising:

inputting data captured by the range cameras, wherein the captured data includes a path of a moving person;

synchronizing data captured by one range camera with data captured by another range camera; and

matching synchronized data from each of the range cameras to calibrate the range cameras and storing the matched synchronized data in a memory of a general-purpose computing device;

computing transformation parameters from the synchronized data; and

expressing data captured by the range cameras in a base coordinate frame using the transformation parameters and without the use of calibration targets or charts.

8. The computer-implemented method of claim 7 , further comprising:

selecting one of the range cameras as a base camera;

selecting as a base coordinate frame a coordinate frame of the base camera; and

measuring a position and orientation between any one of the range cameras and the base camera.

9. The computer-implemented method of claim 7 , wherein synchronizing data further comprises adding a time offset value to the captured data.

10. The computer-implemented method of claim 7 , further comprising:

sampling two data points from the captured data, wherein the first data point is sampled before a time t and the second data point is sampled after the time t; and

linearly interpolating a data point at time t using the first and second data points.

11. The computer-implemented method of claim 7 , further comprising using a least median of squares minimization technique to find data points in the synchronized data that give transformation parameters having a least amount of error.

12. A method for calibrating a first range camera and a second range camera, comprising:

sampling a first data point before a time t and a second data point after time t using the first range camera;

interpolating a data point between the first data point and the second data point to obtain an interpolated data point at time t such that the interpolated data point approximates where a data point would have been seen at time t; and

using the interpolated data point and data sampled by the second range camera to find and store transformation parameters on a general-purpose computing device; and

expressing data captured by the first range camera and the second range camera in a base coordinate frame using the transformation parameters and without the use of calibration targets or charts;

wherein the transformation parameters are used to achieve calibration of the first and second range cameras.

13. The method of claim 12 , further comprising:

the first range camera to capture first camera data and using the second range camera to capture second camera data; and

applying a time offset to the first camera data and the second camera data.

14. The method of claim 12 , wherein the first camera data and the second camera data include a person moving around in a path.

15. The method of claim 13 , further comprising synchronizing the first and second camera data with each other to obtain synchronized data.

16. The method of claim 15 , further comprising performing data matching of the synchronized data to achieve calibration of the first and second range cameras.

17. The method of claim 14 , further comprising:

finding data points in the first and second camera data that represent a minimum error; and

finding transformation parameters using the data points representing the minimum error; and

using the transformation parameters to calibrate the first and second range cameras.

18. The method of claim 17 , further comprising using a least median of squares technique to determine the data points representing the minimum error.