| CPC G06T 7/80 (2017.01) [G06T 7/70 (2017.01); G06T 2207/30196 (2013.01); G06T 2207/30244 (2013.01)] | 18 Claims |
|
1. A camera comprising:
an electronic processor configured to:
obtain an image captured using the camera, the image including a person;
obtain a bounding box in the image, the bounding box bounding the person and having a lowermost vertical coordinate ybottom and an uppermost vertical coordinate ytop;
determine an initial estimation of values of a set of camera parameters Θ;
determine a current estimation of values of the set of camera parameters Θ based on the initial estimation; and
iteratively update the current estimation of values of the set of camera parameters Θ by:
determining feet coordinates (xf, yf) within the bounding box, the feet coordinates (xf, yf) associated with a location of feet of the person, wherein the feet coordinates (xf, yf) are determined based on the current estimation of the values of the set of camera parameters Θ by
placing a human triangle model over the bounding box, the human triangle model having a first vertex l corresponding to a left foot of the person, a second vertex r corresponding to a right foot of the person, and a third vertex v corresponding to a head of the person, wherein the feet coordinates (xf, yf) correspond to a midpoint m between the first vertex l and the second vertex r;
determining a vertical head coordinate yh associated with an estimated vertical location of the head of the person, wherein the vertical head coordinate yh is determined based on the current estimation of the values of the set of camera parameters Θ and the feet coordinates (xf, yf); and
updating the current estimation of values of the set of camera parameters Θ by reducing a difference between the vertical head coordinate yh and the uppermost vertical coordinate ytop of the bounding box.
|