Methods and systems for augmenting depth data from a depth sensor, such as with data from a multiview camera system
Methods of determining the depth of a scene and associated systems are disclosed herein. In some embodiments, a method can include augmenting depth data of a scene captured with a depth sensor with depth data from one or more images of the scene. For example, the method can include capturing image data of the scene with a plurality of cameras. The method can further include generating a point cloud representative of the scene based on the depth data from the depth sensor and identifying a missing region of the point cloud, such as a region occluded from the view of the depth sensor. The method can then include generating depth data for the missing region based on the image data. Finally, the depth data for the missing region can be merged with the depth data from the depth sensor to generate a merged point cloud representative of the scene.
1. A method of determining a depth within a surgical scene, the method comprising:
capturing depth data of anatomy of a patient within the surgical scene undergoing a surgical procedure with a depth sensor;
capturing image data of the anatomy with a plurality of cameras;
generating a point cloud representative of the anatomy based on the depth data from the depth sensor;
identifying a missing region of the point cloud in which the point cloud includes no data or sparse data, wherein the missing region corresponds to a portion of the anatomy that is occluded from the depth sensor;
determining at least one depth value from the point cloud for an area adjacent to the missing region;
processing the image data with a depth processing algorithm to generate depth data for the missing region, wherein processing the image data with the depth processing algorithm includes limiting the depth processing algorithm to determine the depth data for the missing region in a depth range surrounding the at least one depth value; and
merging the depth data for the missing region with the depth data from the depth sensor to generate a merged point cloud representative of the anatomy.
2. The method of claim 1 wherein identifying the missing region of the point cloud includes determining that the missing region of the point cloud has fewer than a predetermined threshold number of data points.
3. The method of claim 1 wherein identifying the missing region of the point cloud includes identifying a hole in the point cloud that is larger than a user-defined threshold.
4. The method of claim 1 wherein the depth data for the missing region has a greater resolution than the depth data captured with the depth sensor.
5. The method of claim 1 wherein the method further comprises generating a three-dimensional mesh representative of the surgical scene based on the merged point cloud.
6. The method of claim 1 wherein the cameras and the depth sensor are rigidly mounted to a common frame and fixed in position relative to one another.
7. The method of claim 6 wherein the cameras are RGB cameras.
8. The method of claim 1 wherein the image data is light field image data.
9. A system for imaging a surgical scene, comprising:
multiple cameras arranged at different positions and orientations relative to the surgical scene and configured to capture image data of anatomy of a patient within the surgical scene undergoing a surgical procedure;
a depth sensor configured to capture depth data of the anatomy; and
a computing device communicatively coupled to the cameras and the depth sensor, wherein the computing device has a memory containing computer-executable instructions and a processor for executing the computer-executable instructions contained in the memory, and wherein the computer-executable instructions, when executed by the processor, cause the processor to:
receive the depth data of the anatomy from the depth sensor;
receive the image data of the anatomy from the cameras;
generate a point cloud representative of the anatomy based on the depth data from the depth sensor;
identify a missing region of the point cloud in which the point cloud includes no data or sparse data, wherein the missing region corresponds to a portion of the anatomy that is occluded from the depth sensor;
determine at least one depth value from the point cloud for an area adjacent to the missing region;
process the image data with a depth processing algorithm to generate depth data for the missing region, wherein the computer-executable instructions, when executed by the processor, further cause the processor to limit the depth processing algorithm to determine the depth data for the missing region in a depth range surrounding the at least one depth value; and
merge the depth data for the missing region with the depth data from the depth sensor to generate a merged point cloud representative of the anatomy.
10. The system of claim 9 wherein the computer-executable instructions, when executed by the processor, cause the processor to identify the missing region of the point cloud by determining that the missing region of the point cloud has fewer than a predetermined threshold number of data points.
11. The system of claim 9 wherein the computer-executable instructions, when executed by the processor, cause the processor to identify the missing region of the point cloud by identifying a hole in the point cloud that is larger than a user-defined threshold.
12. The system of claim 9 wherein the depth data for the missing region has a greater resolution than the depth data captured with the depth sensor.
13. The system of claim 9 wherein the computer-executable instructions, when executed by the processor, further cause the processor to generate a three-dimensional mesh representative of the surgical scene based on the merged point cloud.
14. The system of claim 9 wherein the cameras and the depth sensor are rigidly mounted to a common frame and fixed in position relative to one another.
15. The system of claim 14 wherein the cameras are RGB cameras.
16. A method of generating an output image of a surgical scene, the method comprising:
capturing depth data of the anatomy of a patient within the surgical scene undergoing a surgical procedure with a depth sensor;
capturing light field image data of the anatomy with a plurality of cameras;
generating a point cloud representative of the anatomy based on the depth data from the depth sensor;
identifying a missing region of the point cloud within the field of view of the virtual camera in which the point cloud includes no data or sparse data, wherein the missing region corresponds to a portion of the anatomy that is occluded from the depth sensor;
determining at least one depth value from the point cloud for an area adjacent to the missing region;
processing the light field image data with a depth processing algorithm to generate depth data for the missing region, wherein processing the light field image data with the depth processing algorithm includes limiting the depth processing algorithm to determine the depth data for the missing region in a depth range surrounding the at least one depth value;
merging the depth data for the missing region with the depth data from the depth sensor to generate a merged point cloud representative of the surgical scene;
processing the light field image data and the merged point cloud to synthesize the output image of the surgical scene; and
transmitting the output image to a display for display to a user.
17. The method of claim 16 wherein the output image is from the perspective of a virtual camera having a field of view corresponding to a portion of the surgical scene.
18. The method of claim 17 wherein the plurality of cameras each have a different perspective relative to the surgical scene, and wherein the perspective of the virtual camera is different from any of the perspectives of the cameras.
19. The method of claim 16 wherein the cameras and the depth sensor are rigidly mounted to a common frame and fixed in position relative to one another.
20. The method of claim 16 wherein the cameras are RGB cameras.