Calibrating cameras using human skeleton
Examples are disclosed herein that relate to automatically calibrating cameras based on human detection. One example provides a computing system comprising instructions executable to receive image data comprising depth image data and two-dimensional image data of a space from a camera, detect a person in the space via the image data, determine a skeletal representation for the person via the image data, determine over a period of time a plurality of locations at which a reference point of the skeletal representation is on a ground area in the image data, determine a ground plane of the three-dimensional representation based upon the plurality of locations at which the reference point of the skeletal representation is on the ground area in the image data, and track a location of an object within the space relative to the ground plane.
1. A computing system comprising:
a processor; and
storage comprising instructions executable by the processor to:
receive image data comprising depth image data and two-dimensional image data of a space from a camera,
detect a person in the space via the image data,
determine a skeletal representation for the person via the image data,
determine over a period of time a plurality of locations at which a reference point of the skeletal representation is on a ground area in the image data,
determine a ground plane of the three-dimensional representation based upon the plurality of locations at which the reference point of the skeletal representation is on the ground area in the image data, and
track a location of an object within the space relative to the ground plane.
2. The computing system of claim 1 , wherein the reference point of the skeletal representation is based on a lowest node of the skeletal representation.
3. The computing system of claim 1 , wherein the instructions are further executable to define a convex hull that includes the plurality of locations at which the reference point of the skeletal representation is on the ground area in the image data, map a plurality of points within the convex hull to a three-dimensional representation of the space to form a point cloud, and determine the ground plane based upon the point cloud.
4. The computing system of claim 3 , wherein the instructions are further executable to form the point cloud by mapping a plurality of two-dimensional points within the convex hull as determined from the two-dimensional image data to the three-dimensional representation of the space.
5. The computing system of claim 1 , wherein the instructions are executable to receive image data from a plurality of cameras and to determine the ground plane based upon the image data from the plurality of cameras.
6. The computing system of claim 5 , wherein the instructions are executable to spatially calibrate a pair of cameras of the plurality of cameras based at least upon a skeleton position of the person in an image from a first camera of the pair of cameras and the skeleton position of the person in an image from a second camera of the pair of cameras acquired when the person is located in the overlapping field of view.
7. The computing system of claim 6 , wherein the instructions are executable to spatially calibrate the pair of cameras further based upon the ground plane.
8. A method of tracking an object, the method comprising:
receiving image data comprising depth image data and two-dimensional image data of a space from a camera;
detecting a person in the space via the image data;
determining a skeletal representation for the person via the image data;
determining over a period of time a plurality of locations at which a reference point of the skeletal representation is on a ground area in the image data;
determining a ground plane of the three-dimensional representation based upon the plurality of locations at which the reference point of the skeletal representation is on the ground area in the image data; and
tracking a location of an object within the space relative to the ground plane.
9. The method of claim 8 , further comprising defining a convex hull that includes the plurality of locations at which the reference point of the skeletal representation is on the ground area in the image data, mapping a plurality of points within the convex hull to a three-dimensional representation of the space to form a point cloud, and determining the ground plane based upon the point cloud.
10. The method of claim 8 , further comprising forming the point cloud by mapping a plurality of two-dimensional points within the convex hull as determined from the two-dimensional image data to the three-dimensional representation of the space.
11. The method of claim 8 , further comprising receiving image data from a plurality of cameras and to determine the ground plane based upon the image data from the plurality of cameras.
12. The method of claim 8 , further comprising spatially calibrating a pair of cameras of the plurality of cameras based at least upon a skeleton position of the person in an image from a first camera of the pair of cameras and the skeleton position of the person in an image from a second camera of the pair of cameras acquired when the person is located in the overlapping field of view.
13. The method of claim 8 , further comprising spatially calibrating the pair of cameras further based upon the ground plane.