Find pose of object and transform onto Map
I want to find the pose of a human moving in a simulation world, and transform it in world coordinate and update the occupancy grid map.
I have depth cloud, depth image, laser scan and RGB image sensor data.
The robot moves in the map using AMCL navigation and runs a object detection node with Yolo, it recognizes the human and gives Bounding Box coordinates in Camera Image Frame.
Can anyone give an idea of how to proceed ?