gistfile1.txt

#!/usr/bin/env python

import rospy
import message_filters

from std_msgs.msg import ColorRGBA
from geometry_msgs.msg import Vector3
from jsk_recognition_msgs.msg import ClassificationResult, BoundingBoxArray
from visualization_msgs.msg import MarkerArray, Marker

import numpy as np

def callback(class_msg, bbox_msg):
    rospy.loginfo("found {}".format(zip(class_msg.label_names, class_msg.label_proba)))
    if len(class_msg.label_names) == 0:
        return
    target_names = class_msg.target_names
    msg = MarkerArray()
    for label, name, prob, bbox in zip(class_msg.labels, class_msg.label_names, class_msg.label_proba, bbox_msg.boxes):
        msg.markers.append(Marker(header=class_msg.header,
                                  type=Marker.TEXT_VIEW_FACING,
                                  action=Marker.ADD,
                                  id=np.random.randint(10000),
                                  scale=Vector3(0.1, 0.1, 0.1),
                                  lifetime=rospy.Duration(10),
                                  text="{} ({:.2f})".format(name, prob),
                                  color=ColorRGBA(1.0,1,1.0,1.0),
                                  pose=bbox.pose))
    marker_pub.publish(msg)
    # The callback processing the pairs of numbers that arrived at approximately the same time

rospy.init_node('publish_bbox_label')
class_sub = message_filters.Subscriber('/spot/ncb_provider/class', ClassificationResult)
bbox_sub = message_filters.Subscriber('/spot/ncb_provider/bbox_array', BoundingBoxArray)
marker_pub = rospy.Publisher('/spot/ncb_provider/class_marker', MarkerArray, queue_size=1)

ts = message_filters.ApproximateTimeSynchronizer([class_sub, bbox_sub], 1, 0.1, allow_headerless=True)
ts.registerCallback(callback)
rospy.spin()