我正确测量了这个张量流物体检测模型的FPS吗？

Question

tensorRT优化模型的实测FPS值约为70fps或约13ms处理时间。但是，这些帧似乎比实际视频要慢一些，因为我可以清楚地观察到输出帧中的人的运动比原始视频要慢。

因此，我认为我测量fps的方式可能有问题。

这里是代码的重要部分：

times = []

# Detection
with tf.device('/gpu:0'):
    with detection_graph.as_default():
        with tf.Session(graph=detection_graph) as sess:
            while True:
                # Read frame from camera
                ret, image_np = cap.read()
                if not ret:
                    print('cap.read() did not return image\n')
                    break
                # Expand dimensions since the model expects images to have shape: [1, None, None, 3]
                image_np_expanded = np.expand_dims(image_np, axis=0)
                # Extract image tensor
                image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
                # Extract detection boxes
                boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
                # Extract detection scores
                scores = detection_graph.get_tensor_by_name('detection_scores:0')
                # Extract detection classes
                classes = detection_graph.get_tensor_by_name('detection_classes:0')
                # Extract number of detectionsd
                num_detections = detection_graph.get_tensor_by_name(
                  'num_detections:0')

                # Actual detection.
                t1 = time()

                (boxes, scores, classes, num_detections) = sess.run(
                  [boxes, scores, classes, num_detections],
                  feed_dict={image_tensor: image_np_expanded})

                t2 = time()
                times.append(t2-t1)

                # Visualization of the results of a detection.
                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,
                  use_normalized_coordinates=True,
                  line_thickness=8)

                # Display output
                image_np = cv2.putText(image_np, "Time: {:,.2f}ms | FPS: {:.2f}fps".format(times[-1]*1000, 1/times[-1]),
                                       (0, 20), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (255, 0, 255), 2)

                cv2.imshow('object detection', cv2.resize(image_np, (800, 600)))
                if cv2.waitKey(1) & 0xFF == ord('q'):
                  break


# cap.release()
cv2.destroyAllWindows()

if len(times) > 0:
    print('Average FPS: {:,.2f}fps'.format(1/(sum(times)/len(times))))

Answer 1

我看不到这些代码有任何反面。