To re-use the simple 3D pose estimation form find_object_2d node, you need correspnding camera_info and depth image. To generate the depth image from the lidar, you need proper TF tree and the node
pointcloud_to_depthimage could be used to generate the depth image.
cheers,
Mathieu