Method and apparatus for computing error-bounded position and orientation of panoramic cameras in real-world environments
View Patent ↗Techniques for computing error-bounded position and orientation (pose) of a panoramic camera in real-world environments. Such environments may include large interior spaces (e.g., buildings) A space may include multiple rooms. For example, a technique for capturing images associated with an environment includes the following steps/operations. First, respective placements of fiducials in the environment are determined so as to satisfy at least one constraint. Images are captured, with an image capture device (e.g., camera), associated with the environment with the fiducials placed therein. A pose estimation of the image capture device is then determined based on projections of the fiducials in the captured images. The pose estimation may be optimized so as to obtain an optimal pose per image. Also, the fiducial placements may be optimized so as to obtain optimal fiducial placements. Then at least one constraint may include a constraint associated with the number of visible fiducials, a constraint associated with a distance from a viewpoint to a fiducial, and/or a constraint associated with an angle subtended by pairs of fiducials.
1. A method of estimating pose for a panoramic camera in an environment, comprising the steps of:
determining placements of fiducials in the environment, wherein fiducial placement is determined in accordance with an automated planning algorithm;
determining an error-bounded pose of the camera based on projections of the fiducials in images of the environment captured by the camera; and
refining the fiducial placements so as to obtain fiducial placements that provide bounded error pose estimates.
2. The method of claim 1 , wherein the step of determining fiducial placements further comprises the step of satisfying at least one constraint.
3. The method of claim 2 , wherein the at least one constraint comprises a constraint associated with the number of visible fiducials.
4. The method of claim 2 , wherein the at least one constraint comprises a constraint associated with a distance from a viewpoint to a fiducial.
5. The method of claim 2 , wherein the at least one constraint comprises a constraint associated with an angle subtended by vectors from a viewpoint to pairs of fiducials.
6. The method of claim 1 , wherein the step of determining fiducial placements in the environment further comprises:
creating an initial list of fiducials that satisfy at least one constraint;
prioritizing the fiducials in a placement list; and
iteratively removing fiducials from the placement list, based on priority, until no more fiducials can be removed while still satisfying the at least one constraint, wherein the placement list, resulting from the iterative removal of fiducials, represents the fiducials to be placed.
7. The method of claim 6 , wherein the step of creating an initial list of fiducials comprises the steps of:
obtaining a floor plan of the environment;
dividing the floor plan into convex regions; and
positioning a number of fiducials so as to satisfy the at least one constraint within each convex region.
8. The method of claim 6 , wherein the step of prioritizing the fiducials further comprises the steps of:
dividing the fiducials into two or more categories based on their general positioning within the environment;
prioritizing the fiducials within each category based on how effectively they satisfy the at least one constraint; and
prioritizing the fiducial categories based on their general effectiveness as fiducial locations.
9. The method of claim 1 , wherein the step of determining an error-bounded pose of the camera further comprises:
tracking the fiducials based on their projections in the captured images; and
triangulating the pose of the camera relative to the tracked fiducials.
10. The method of claim 1 , further comprising the step of adapting the pose determination so as to obtain an improved pose per image by refining pose and overall fiducial positions in the images.
11. An apparatus for estimating pose for a panoramic camera in an environment, comprising:
a memory; and
at least one processor coupled to the memory and operative to: (i) determine placements of fiducials in the environment, wherein fiducial placement is determined in accordance with an automated planning algorithm; (ii) determine an error-bounded pose of the camera based on projections of the fiducials in images of the environment captured by the camera; and (iii) refine the fiducial placements so as to obtain fiducial placements that provide bounded error pose estimates.
12. The apparatus of claim 11 , wherein the operation of determining fiducial placements further comprises the step of satisfying at least one constraint.
13. The apparatus of claim 12 , wherein the at least one constraint comprises a constraint associated with the number of visible fiducials.
14. The apparatus of claim 12 , wherein the at least one constraint comprises a constraint associated with a distance from a viewpoint to a fiducial.
15. The apparatus of claim 12 , wherein the at least one constraint comprises a constraint associated with an angle subtended by vectors from a viewpoint to pairs of fiducials.
16. The apparatus of claim 11 , wherein the operation of determining fiducial placements in the environment further comprises:
creating an initial list of fiducials that satisfy at least one constraint;
prioritizing the fiducials in a placement list; and
iteratively removing fiducials from the placement list, based on priority, until no more fiducials can be removed while still satisfying the at least one constraint, wherein the placement list, resulting from the iterative removal of fiducials, represents the fiducials to be placed.
17. The apparatus of claim 16 , wherein the operation of creating an initial list of fiducials comprises the steps of:
obtaining a floor plan of the environment;
dividing the floor plan into convex regions; and
positioning a number of fiducials so as to satisfy the at least one constraint within each convex region.
18. The apparatus of claim 16 , wherein the operation of prioritizing the fiducials further comprises the steps of:
dividing the fiducials into two or more categories based on their general positioning within the environment;
prioritizing the fiducials within each category based on how effectively they satisfy the at least one constraint; and
prioritizing the fiducial categories based on their general effectiveness as fiducial locations.
19. The apparatus of claim 11 , wherein the operation of determining an error-bounded pose of the camera further comprises:
tracking the fiducials based on their projections in the captured images; and
triangulating the pose of the camera relative to the tracked fiducials.
20. The apparatus of claim 11 , wherein the at least one processor is further operative to adapt the pose determination so as to obtain an improved pose per image by refining pose and overall fiducial positions in the images.
21. An article of manufacture for estimating pose for a panoramic camera in an environment, comprising a non-transitory machine readable medium containing one or more programs which when executed implement the steps of:
determining placements of fiducials in the environment, wherein fiducial placement is determined in accordance with an automated planning algorithm;
determining an error-bounded pose of the camera based on projections of the fiducials in images of the environment captured by the camera; and
refining the fiducial placements so as to obtain fiducial placements that provide bounded error pose estimates.