Print class name from tensorflow object detection api

Question

PLEASE NOTE: I have tried other solutions accross the web and didnt find the working result. I am detecting objects from live feed using tensorflow object detection api and opencv. It is working fine and showing the boxes on the screen. Now i want to only print the detected object on the terminal. I Have tried various solutions and couldnt find anything which worked. I am new to this. Here's my code:

while True:
    ret, frame = cap.read()
image_np = np.array(frame)

input_tensor = tf.convert_to_tensor(
    np.expand_dims(image_np, 0), dtype=tf.float32)
detections = detect_fn(input_tensor)

num_detections = int(detections.pop('num_detections'))
detections = {key: value[0, :num_detections].numpy()
              for key, value in detections.items()}
detections['num_detections'] = num_detections

# detection_classes should be ints.
detections['detection_classes'] = detections['detection_classes'].astype(
    np.int64)

label_id_offset = 1
image_np_with_detections = image_np.copy()


viz_utils.visualize_boxes_and_labels_on_image_array(
    image_np_with_detections,
    detections['detection_boxes'],
    detections['detection_classes']+label_id_offset,
    detections['detection_scores'],
    category_index,
    use_normalized_coordinates=True,
    max_boxes_to_draw=5,
    min_score_thresh=.5,
    agnostic_mode=False)

cv2.imshow('object detection',  cv2.resize(
    image_np_with_detections, (800, 600)))

I tried using:

print(detections['detection_classes'])

But that just prints a set of values.Any help would be appreciated

Answer 1

category_index is a dictionary with your label names. Index it with detections['detection_classes'] to get the name of the categories. Something like this, but this won't work because I'm writing it directly on here without testing:

label_names = [i['name'] for i in category_index.items()]
label_names = np.array(label_names)

print(label_names[detections['detection_classes'].numpy()])