Adaptive model updates for dynamic and static scenes
In one embodiment, a computing system may update a first 3D model of a region of an environment based on comparisons between the first 3D model and first depth measurements of the region generated during a first time period. The computing system may determine that the region is static by comparing the first 3D model to second depth measurements of the region generated during a second time period. The computing system may in response to determining that the region is static, detect whether the region changed after the second time period based on comparisons between a second 3D model of the region and third depth measurements of the region generated after the second time period, the second 3D model having a lower resolution than the first 3D model. The computing system may in response to detecting a change in the region, update the first 3D model of the region.
1. A method comprising, by a computing system:
updating a first 3D model of a region of an environment based on comparisons between the first 3D model and first depth measurements of the region of the environment generated during a first time period;
determining that the region of the environment is static by comparing the first 3D model to second depth measurements of the region of the environment generated during a second time period;
in response to the determining that the region of the environment is static, detecting whether the region of the environment changed after the second time period based on comparisons between a second 3D model of the region of the environment and third depth measurements of the region of the environment generated after the second time period; and
in response to detecting a change in the region of the environment, updating the first 3D model of the region of the environment.
2. The method of claim 1 , wherein the first 3D model comprises a plurality of voxels.
3. The method of claim 2 , wherein the second 3D model comprises a set of the plurality of voxels of the first 3D model.
4. The method of claim 1 , wherein the second 3D model comprises a plurality of voxel occupancy values.
5. The method of claim 1 , further comprising:
identifying one or more objects using the first 3D model to detect objects m the environment; and
assigning one or more characteristics to the identified one or more objects based on a respective object.
6. The method of claim 5 , further comprising:
segmenting the environment into a plurality of regions based on the one or more characteristics assigned to the identified one or more objects.
7. The method of claim 5 , further comprising:
tracking the one or more identified objects based on the first 3D model, wherein updates to the first 3D model updates a position of the one or more identified objects.
8. The method of claim 1 , further comprising:
segmenting the environment into a plurality of regions based on a predetermined segmentation pattern for the environment.
9. The method of claim 1 , wherein one or more characteristics are associated with the region of the environment, and wherein the detecting whether the region of the environment changed occurs after a third time period based on the one or more characteristics associated with the region of the environment.
10. The method of claim 1 , wherein the detecting whether the region of the environment changed comprises comparing the third depth measurements of the region of the environment to a plurality of voxels of the second 3D model of the region of the environment.
11. The method of claim 1 , wherein the detecting whether the region of the environment changed comprises comparing the third depth measurements of the region of the environment to a plurality of voxel occupancy values of the second 3D model of the region of the environment.
12. The method of claim 1 , wherein the detecting whether the region of the environment changed comprises determining whether the change between the second 3D model of the region of the environment and the third depth measurements of the region of the environment generated after the second time period exceeds a threshold change.
13. The method of claim 12 , wherein the threshold change comprises one of a threshold number of voxels or a threshold number of voxel occupancy values.
14. The method of claim 1 , further comprising: labeling one or more other regions of the first 3D model based on a frequency of previous updates to the one or more other regions of the first 3D model, wherein the detecting whether the region of the environment changed occurs after a third time period based on a label of the region of the environment.
15. The method of claim 1 , further comprising:
detecting whether a second region changed after a third time period based on comparisons between the second 3D model of the second region and fourth depth measurements of the second region; and
in response to detecting a change in the second region falls below a threshold change, skipping an update to the first 3D model of the second region.
16. The method of claim 1 , wherein the second 3D model has a lower resolution than the first 3D model.
17. One or more computer-readable non-transitory storage media embodying software that is operable when executed by one or more processors to:
update a first 3D model of a region of an environment based on comparisons between the first 3D model and first depth measurements of the region of the environment generated during a first time period;
determine that the region of the environment is static by comparing the first 3D model to second depth measurements of the region of the environment generated during a second time period;
in response to determining that the region of the environment is static, detect whether the region of the environment changed after the second time period based on comparisons between a second 3D model of the region of the environment and third depth measurements of the region of the environment generated after the second time period; and
in response to detecting a change in the region of the environment, update the first 3D model of the region of the environment.
18. The one or more computer-readable non-transitory storage media of claim 17 , wherein the one or more computer-readable non-transitory storage media is further operable when executed to:
identify one or more objects using the first 3D model to detect objects m the environment; and
assign one or more characteristics to the identified one or more objects based on a respective object.
19. A system comprising:
one or more processors; and
one or more computer-readable non-transitory storage media coupled to the one or more processors and comprising instructions operable when executed by the one or more processors to cause the system to:
update a first 3D model of a region of an environment based on comparisons between the first 3D model and first depth measurements of the region of the environment generated during a first time period;
determine that the region of the environment is static by comparing the first 3D model to second depth measurements of the region of the environment generated during a second time period;
in response to determining that the region of the environment is static, detect whether the region of the environment changed after the second time period based on comparisons between a second 3D model of the region of the environment and third depth measurements of the region of the environment generated after the second time period; and
in response to detecting a change in the region of the environment, update the first 3D model of the region of the environment.
20. The system of claim 19 , wherein the instructions are further executable by the one or more processors to:
identify one or more objects using the first 3D model to detect objects m the environment; and
assign one or more characteristics to the identified one or more objects based on a respective object.