Aligning data sets based on identified fiducial markers
Techniques are disclosed for aligning fiducial markers that commonly exist in each of multiple different N-dimensional (N-D) data sets. Notably, the N-D data sets are at least three-dimensional (3D) data sets. A first set and a second set of N-D data are accessed. A set of one or more fiducial markers that commonly exist in both those sets are identified. Based on the fiducial markers, one or more transformations are performed to align the two sets. Performing this alignment process results in at least a selected number of the common fiducial markers that exist in the two sets being within a threshold alignment relative to one another.
1. A method for aligning fiducial markers that commonly exist in each of multiple different N-dimensional (N-D) data sets, wherein the N-D data sets are at least three-dimensional (3D) data sets, said method comprising:
accessing a first set of N-D data;
accessing a second set of N-D data;
identifying a set of one or more fiducial markers that commonly exist in both the first set of N-D data and the second set of N-D data; and
based on the identified set of one or more fiducial markers, perform one or more transformations to the first set of N-D data and/or the second set of N-D data to align the first set of N-D data with the second set of N-D data, wherein said aligning results in at least a selected number of the common fiducial markers that exist in the second set of N-D data being within a threshold alignment relative to the corresponding common fiducial markers that exist in the first set of N-D data.
2. The method of claim 1 , wherein the set of one or more fiducial markers includes at least 10 fiducial markers.
3. The method of claim 1 , wherein the one or more transformations includes one or more of a rotation transformation, a scaling transformation, and a translation transformation.
4. The method of claim 1 , wherein the method further includes:
subsequent to aligning the first set of N-D data with the second set of N-D data, attempting to identify one or more differences that exist between the aligned first set of N-D data and the aligned second set of N-D data.
5. The method of claim 4 , wherein the method further includes:
in response to identifying the one or more differences, attempting to classify the one or more differences to determine what the one or more differences represent.
6. The method of claim 1 , wherein the first set of N-D data and the second set of N-D data represent a common entity, and wherein the first set of N-D data and the second set of N-D data are time series data such that the first set of N-D data has a first timestamp and the second set of N-D data has a second timestamp.
7. The method of claim 1 , wherein the first set of N-D data is generated by a first camera of a first modality, and the second set of N-D data is generated by a second camera of a second modality.
8. The method of claim 1 , wherein the first set of N-D data and the second set of N-D data are generated by different sensor types.
9. The method of claim 1 , wherein a machine learning algorithm identifies the set of one or more fiducial markers.
10. The method of claim 1 , wherein at least one fiducial marker in the set of one or more fiducial markers is identified via user input.
11. A computer system configured to align fiducial markers that commonly exist in each of multiple different three-dimensional (3D data sets, said computer system comprising:
one or more processors; and
one or more computer-readable hardware storage devices that store instructions that are executable by the one or more processors to cause the computer system to at least:
access a first set of 3D data;
access a second set of 3D data;
identify a set of one or more fiducial markers that commonly exist in both the first set of 3D data and the second set of 3D data; and
based on the identified set of one or more fiducial markers, perform one or more transformations to the first set of 3D data and/or the second set of 3D data to align the first set of 3D data with the second set of 3D data, wherein said aligning results in at least a selected number of the common fiducial markers that exist in the second set of 3D data being within a threshold alignment relative to the corresponding common fiducial markers that exist in the first set of 3D data.
12. The computer system of claim 11 , wherein the first set of 3D data is generated by one or more of a wearable device, a depth sensor, or an infrared (IR) sensor.
13. The computer system of claim 11 , wherein the first set of 3-D data is 3D point cloud data or surface reconstruction mesh data or depth map data.
14. The computer system of claim 11 , wherein the computer system refrains from filtering any data from the first set of 3D data and the second set of 3D data.
15. The computer system of claim 12 , wherein, as a result of refraining from filtering the first set of 3D data and the second set of 3D data, one or more differences exist between the first set of 3D data and the second set of 3D data.
16. The computer system of claim 15 , wherein the instructions are further executable to cause the computer system to at least:
identify and classify the differences to determine what said differences potentially represent.
17. The computer system of claim 11 , wherein identifying the set of one or more fiducial markers that commonly exist in both the first set of 3D data and the second set of 3D data includes performing one of:
object segmentation to identify features in the first set of 3D and the second set of 3D data; or
voxel intensity analysis; or
invariant feature detection.
18. The computer system of claim 11 , wherein the instructions are further executable to cause the computer system to at least:
subsequent to aligning the first set of 3D data with the second set of 3D data, identifying one or more differences that exist between the first set of 3D data and the second set of 3D data;
visually displaying the first set of 3D data and the second set of 3D data; and
visually emphasizing the one or more differences.
19. A computer system configured to align fiducial markers that commonly exist in each of multiple different three-dimensional (3D data sets, said computer system comprising:
one or more processors; and
one or more computer-readable hardware storage devices that store instructions that are executable by the one or more processors to cause the computer system to at least:
access a first set of 3D data;
access a second set of 3D data;
identify a set of one or more fiducial markers that commonly exist in both the first set of 3D data and the second set of 3D data;
based on the identified set of one or more fiducial markers, perform one or more transformations to the first set of 3D data and/or the second set of 3D data to align the first set of 3D data with the second set of 3D data, wherein said aligning results in at least a selected number of the common fiducial markers that exist in the second set of 3D data being within a threshold alignment relative to the corresponding common fiducial markers that exist in the first set of 3D data; and
subsequent to aligning the first set of 3D data with the second set of 3D data, identify one or more differences that exist between the first set of 3D data and the second set of 3D data.
20. The computer system of claim 19 , wherein the instructions are further executable to cause the computer system to at least:
attempt to classify the identified one or more differences.