Method and apparatus for jointly calibrating external parameters of multiple cameras, device and medium

1. A method for jointly calibrating external parameters of multiple cameras, the method comprising:

determining an overlapped area of angles of view of cameras, and obtaining a two-dimensional (2D) verification point set in an image of the overlapped area, the 2D verification point set comprising at least one verification point;

performing external parameter calibration for each camera separately to obtain a current external parameter of each camera;

calculating coordinates of a three-dimensional (3D) point corresponding to each verification point in a world coordinate system using the current external parameter of each camera separately, and calculating a loss function based on the coordinates of the 3D point, the loss function being used to measure an overlay error of calibration of the cameras under the current external parameter; and

performing joint calibration based on the loss function to obtain a target external parameter of each camera.

2. The method according to claim 1 , wherein the performing joint calibration based on the loss function to obtain a target external parameter of each cameras, comprises:

executing repeatedly following operations until a number of executions meets a preset threshold, and using an external parameter of an cameras corresponding to a loss function whose value satisfies a preset condition in loss functions obtained by calculation as the target external parameter of the camera:

performing external parameter calibration for each camera separately to obtain the current external parameter of each camera; and

calculating the coordinates of the 3D point corresponding to each verification point in the world coordinate system using the current external parameter of each camera separately, and calculating the loss function based on the coordinates of the 3D point.

3. The method according to claim 1 , wherein calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, and calculating a variance based on coordinates of a 3D point of the current verification point obtained by calculation corresponding to each camera; and

summing variances obtained by calculation corresponding to verification points, and using a sum value as the loss function.

4. The method according to claim 3 , wherein calculating the loss function based on the coordinates of the 3D point, further comprises:

determining a minimum distance value between any two 3D points in 3D points of the current verification point obtained by calculation corresponding to each camera; and

adding the loss function to the distance value corresponding to each verification point obtained by calculation, on the basis of the loss function obtained by calculation.

5. The method according to claim 1 , wherein calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, determining a distance between a 3D point of the current verification point obtained by calculation corresponding to each camera and the true value based on a truth value of the current verification point in the world coordinate system, and calculating a mean of the distance;

calculating a variance based on coordinates of the 3D point of the current verification point obtained by calculation corresponding to each camera combining the truth value;

weighting and summing the variance and the mean obtained by calculation corresponding to the current verification point; and

summing results of the weighting and summing obtained by calculation corresponding to verification points, and using a sum value as the loss function.

6. The method according to claim 1 , wherein the performing external parameter calibration for each camera separately to obtain the current external parameter of each camera, comprises:

performing external parameter calibration for each camera separately using a closest point matching method and a minimized reprojection error method to obtain the current external parameter of each camera.

7. An apparatus for jointly calibrating external parameters of multiple cameras, the apparatus comprising:

at least one processor; and

a memory storing instructions, wherein the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:

determining an overlapped area of angles of view of cameras, and obtaining a two-dimensional (2D) verification point set in an image of the overlapped area, the 2D verification point set comprising at least one verification point;

performing external parameter calibration for each camera separately to obtain a current external parameter of each cameras;

calculating a coordinate of a three-dimensional (3D) point corresponding to each verification point in a world coordinate system using the current external parameter of each cameras separately, and calculating a loss function based on the coordinates of the 3D point, the loss function being used to measure an overlay error of calibration of the cameras under the current external parameter; and

performing joint calibration based on the loss function to obtain a target external parameter of each camera.

8. The apparatus according to claim 7 , wherein the performing joint calibration based on the loss function to obtain a target external parameter of each cameras, comprises:

executing repeatedly following operations until a number of executions meets a preset threshold, and using an external parameter of an cameras corresponding to a loss function whose value satisfies a preset condition in loss functions obtained by calculation as the target external parameter of the camera:

performing external parameter calibration for each camera separately to obtain the current external parameters of each camera; and

calculating the coordinates of the 3D point corresponding to each verification point in the world coordinate system using the current external parameter of each camera separately, and calculating the loss function based on the coordinates of the 3D point.

9. The apparatus according to claim 7 , wherein

calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, and calculating a variance based on coordinates of a 3D point of the current verification point obtained by calculation corresponding to each camera; and summing variances obtained by calculation corresponding to verification points, and using a sum value as the loss function.

10. The apparatus according to claim 9 , wherein calculating the loss function based on the coordinates of the 3D point, further comprises:

determining a minimum distance value between any two 3D points in 3D points of the current verification point obtained by calculation corresponding to each camera; and

adding the loss function to the distance value corresponding to each verification point obtained by calculation, on the basis of the loss function obtained by calculation.

11. The apparatus according to claim 7 , wherein calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, determining a distance between a 3D point of the current verification point obtained by calculation corresponding to each camera and the true value based on a truth value of the current verification point in the world coordinate system, and calculate a mean of the distance;

calculating a variance based on coordinates of the 3D point of the current verification point obtained by calculation corresponding to each camera combining the truth value;

weighting and summing the variance and the mean obtained by calculation corresponding to the current verification point; and

summing results of the weighting and summing obtained by calculation corresponding to verification points, and using a sum value as the loss function.

12. The apparatus according to claim 7 , wherein the performing external parameter calibration for each camera separately to obtain the current external parameter of each camera, comprises:

performing external parameter calibration for each camera separately using a closest point matching method and a minimized reprojection error method to obtain the current external parameter of each camera.

13. A non-transitory computer-readable storage medium, storing computer instructions, the computer instructions, when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:

determining an overlapped area of angles of view of cameras, and obtaining a two-dimensional (2D) verification point set in an image of the overlapped area, the 2D verification point set comprising at least one verification point;

performing external parameter calibration for each camera separately to obtain a current external parameter of each cameras;

calculating a coordinate of a three-dimensional (3D) point corresponding to each verification point in a world coordinate system using the current external parameter of each cameras separately, and calculating a loss function based on the coordinates of the 3D point, the loss function being used to measure an overlay error of calibration of the cameras under the current external parameter; and

performing joint calibration based on the loss function to obtain a target external parameter of each camera.

14. The non-transitory computer-readable storage medium according to claim 13 , wherein the performing joint calibration based on the loss function to obtain a target external parameter of each cameras, comprises:

executing repeatedly following operations until a number of executions meets a preset threshold, and using an external parameter of an cameras corresponding to a loss function whose value satisfies a preset condition in loss functions obtained by calculation as the target external parameter of the camera:

performing external parameter calibration for each camera separately to obtain the current external parameters of each camera; and

calculating the coordinates of the 3D point corresponding to each verification point in the world coordinate system using the current external parameter of each camera separately, and calculating the loss function based on the coordinates of the 3D point.

15. The non-transitory computer-readable storage medium according to claim 13 , wherein

calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, and calculating a variance based on coordinates of a 3D point of the current verification point obtained by calculation corresponding to each camera; and summing variances obtained by calculation corresponding to verification points, and using a sum value as the loss function.

16. The non-transitory computer-readable storage medium according to claim 15 , wherein calculating the loss function based on the coordinates of the 3D point, further comprises:

determining a minimum distance value between any two 3D points in 3D points of the current verification point obtained by calculation corresponding to each camera; and

adding the loss function to the distance value corresponding to each verification point obtained by calculation, on the basis of the loss function obtained by calculation.

17. The non-transitory computer-readable storage medium according to claim 13 , wherein calculating the loss function based on the coordinates of the 3D point, comprises:

using any verification point as a current verification point, determining a distance between a 3D point of the current verification point obtained by calculation corresponding to each camera and the true value based on a truth value of the current verification point in the world coordinate system, and calculate a mean of the distance;

calculating a variance based on coordinates of the 3D point of the current verification point obtained by calculation corresponding to each camera combining the truth value;

weighting and summing the variance and the mean obtained by calculation corresponding to the current verification point; and

summing results of the weighting and summing obtained by calculation corresponding to verification points, and using a sum value as the loss function.

18. The non-transitory computer-readable storage medium according to claim 13 , wherein the performing external parameter calibration for each camera separately to obtain the current external parameter of each camera, comprises:

performing external parameter calibration for each camera separately using a closest point matching method and a minimized reprojection error method to obtain the current external parameter of each camera.