CODEWITHSUNDEEP
opencvtensorflowobject-detection-api
Aravind Swamy
Aravind Swamy

Reputation: 1

Cannot feed value of shape (480, 640, 3) for Tensor 'image_tensor:0', which has shape '(?, ?, ?, 3)'

I'm trying to run my code for a "Campus Building Detector" and I'm using Tensorflow's object detection api with faster_rcnn_inception_v2 as model. I have trained the network for 7000 times (took 12Hrs) and aborted as it has more number of iterations (I have 900000) and now when I'm trying to run the code, I get the following error:

Cannot feed value of shape (480, 640, 3) for Tensor 'image_tensor:0', which has shape '(?, ?, ?, 3)'

I'm using anaconda, Jupiter notebook, Python v3.6.8, Tensorflow v1.13.1

CODE:

import cv2
cap = cv2.VideoCapture(0)
try:
    with detection_graph.as_default():
        with tf.Session() as sess:
                # Get handles to input and output tensors
                ops = tf.get_default_graph().get_operations()
                all_tensor_names = {output.name for op in ops for output in op.outputs}
                tensor_dict = {}
                for key in [
                  'num_detections', 'detection_boxes', 'detection_scores',
                  'detection_classes', 'detection_masks'
                ]:
                    tensor_name = key + ':0'
                    if tensor_name in all_tensor_names:
                        tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(
                      tensor_name)

                while True:
                    ret, image_np = cap.read()
                    # Expand dimensions since the model expects images to have shape: [1, None, None, 3]
                    image_np_expanded = np.expand_dims(image_np, axis=0)
                    # Actual detection.
                    output_dict = run_inference_for_single_image(image_np, detection_graph)
                    # Visualization of the results of a detection.
                    vis_util.visualize_boxes_and_labels_on_image_array(
                        image_np,
                        output_dict['detection_boxes'],
                        output_dict['detection_classes'],
                        output_dict['detection_scores'],
                        category_index,
                        instance_masks=output_dict.get('detection_masks'),
                        use_normalized_coordinates=True,
                        line_thickness=8)
                    cv2.imshow('object_detection', cv2.resize(image_np, (800, 600)))
                    if cv2.waitKey(25) & 0xFF == ord('q'):
                        cap.release()
                        cv2.destroyAllWindows()
                        break
except Exception as e:
    print(e)
    cap.release()

Thanks in advance.

Upvotes: 0

Views: 1541

Answers (1)

Danny Fang
Danny Fang

Reputation: 4071

The function run_inference_for_single_image expects input of images as in batches (four dimensions), so the line below is trying to expand the image in three dimensions into four,

image_np_expanded = np.expand_dims(image_np, axis=0)

You just need to change the line 

output_dict = run_inference_for_single_image(image_np, detection_graph)

into 

output_dict = run_inference_for_single_image(image_np_expanded, detection_graph)

That will solve the problem.

Upvotes: 1

Related Questions