Skip to content

Instantly share code, notes, and snippets.

@MrDavidYu
Created August 26, 2018 19:41
Show Gist options
  • Star 3 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save MrDavidYu/2933458dbf7600bea56bfba105122794 to your computer and use it in GitHub Desktop.
Save MrDavidYu/2933458dbf7600bea56bfba105122794 to your computer and use it in GitHub Desktop.
with detection_graph.as_default():
with tf.Session(graph=detection_graph) as sess:
image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
num_detections = detection_graph.get_tensor_by_name('num_detections:0')
for image_path in TEST_IMAGE_PATHS:
image = Image.open(image_path)
image_np = load_image_into_numpy_array(image)
image_np_expanded = np.expand_dims(image_np, axis=0)
(boxes, scores, classes, num) = sess.run(
[detection_boxes, detection_scores, detection_classes, num_detections],
feed_dict={image_tensor: image_np_expanded})
vis_util.visualize_boxes_and_labels_on_image_array(
image_np,
np.squeeze(boxes),
np.squeeze(classes).astype(np.int32),
np.squeeze(scores),
category_index,
max_boxes_to_draw=200, # Modify depending on how many objects are likely to be in a given image
min_score_thresh=0.6, # Adjust accordingly, this will determine your detection threshold
use_normalized_coordinates=True,
line_thickness=2)
plt.figure(figsize=IMAGE_SIZE)
plt.imshow(image_np)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment